{"campaigns":{"sync":{"identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","manifestSha256":"b0283b88aab462e69f3063d4af6aedb1a4f4ff4473cb9e05282798b1a45c3d59","manifestPath":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/campaign.json","definition":{"version":1,"base":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.0056,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"runs/20m","cpu_threads":8},"decentralized":null},"grid":{"beta1":[0.9,0.95],"beta2":[0.9,0.98,0.999],"lr":[0.008,0.006,0.004,0.003,0.002],"seeds":[42,43,44]},"horizons":[{"horizon":20,"epochs":40,"tokens":408027136,"steps":3113,"checkpoint_epoch":35,"checkpoint_tokens":357040128,"stable_end":367224422.40000004,"retained_checkpoint_epoch":35},{"horizon":40,"epochs":80,"tokens":816185344,"steps":6227,"checkpoint_epoch":72,"checkpoint_tokens":734527488,"stable_end":734566809.6,"retained_checkpoint_epoch":72},{"horizon":80,"epochs":160,"tokens":1632239616,"steps":12453,"checkpoint_epoch":143,"checkpoint_tokens":1458831360,"stable_end":1469015654.4,"retained_checkpoint_epoch":143},{"horizon":120,"epochs":240,"tokens":2448424960,"steps":18680,"checkpoint_epoch":216,"checkpoint_tokens":2203582464,"stable_end":2203582464.0,"retained_checkpoint_epoch":216},{"horizon":160,"epochs":320,"tokens":3264610304,"steps":24907,"checkpoint_epoch":288,"checkpoint_tokens":2938109952,"stable_end":2938149273.6,"retained_checkpoint_epoch":null}],"code":{"scripts":{"common.py":"dbebcb01399f24482e802fdcb2cfe2c5fa62ac313118ff561838dae09fc3d350","sweep.py":"a9b73783495fffcd6338ad911f7d62c13cd8a564f01366f813b8a5f54b920f3a","worker.py":"35052d7ee8f3f20e58aae9dbbec0405c0ae1eda18d4fad02a2b30c00e2e87865","launcher.sh":"94a4774a94ef2d50288f7f2c7b9a06c70a210e0ca139aceb99fbcb9e1e0e6d0a"},"package":"0ef51b878b33c5375aa24d9c965feb590cfaa3411e1e62d32d15b0bb386909c1","dependencies":{"pyproject.toml":"6194efd1e3b255670aa8a9ee57d9585a5aa5207ff68fe81c4d38615322c6309f","uv.lock":"11fe47e1e133585bfc4fc83fd2498320dfcb349655e565a166489c668447cb6f"},"build_inputs":{"README.md":"5049d9bfca2069db79a6371a8e026d9ef4fea8d9061a644220e0138477e42bb4",".python-version":"7b55f8e67b5623c4bef3fa691288da9437d79d3aba156de48d481db32ac7d16d"}},"cache_identity":"7618ae8749071f012cd3e3e27f96480ea18c5da5dca59cf90330f35beaa4da1c","validation_tokens":197411295,"max_concurrent":48,"walltime_hours":2,"account":"naiss2026-3-205-gpu","partition":"gpu","gpu":"nvidia_gh200_120gb:1","pilot":"b1-0.9-b2-0.98-lr-0.004-s42","selection":"mean-three-seed-full-validation","retention":"all-final-weights-winner-states-needed-parents"},"completedAt":1789428516.4713566,"retention":[{"time":1789423156.934628,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.safetensors"]},{"time":1789424467.779437,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.safetensors"]},{"time":1789426186.31023,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.safetensors"]},{"time":1789427352.5483658,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.safetensors"]},{"time":1789428516.4331808,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt"]}],"attempts":[{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-f415dc1dcc-a1","created_at":1789420631.5203996,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-f415dc1dcc-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450844","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-498332c166-a1","created_at":1789420631.6354978,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-498332c166-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450845","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-31e6b398d5-a1","created_at":1789420631.6964972,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-31e6b398d5-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450846","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-ac5caa08db-a1","created_at":1789420631.7890809,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-ac5caa08db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450847","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-46c6775a30-a1","created_at":1789420631.8616297,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-46c6775a30-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450848","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-7d4714319e-a1","created_at":1789420631.8917813,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-7d4714319e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450849","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-367f579315-a1","created_at":1789420631.9266615,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-367f579315-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450850","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-b95fab87d6-a1","created_at":1789420632.0014682,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-b95fab87d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450851","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-e86c77c879-a1","created_at":1789420632.3059537,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-e86c77c879-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450852","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-8a5a73b5f9-a1","created_at":1789420632.3860395,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450853","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-5f9d870038-a1","created_at":1789420632.4336207,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450854","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-66563544e3-a1","created_at":1789420632.480242,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450855","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-14813e27f8-a1","created_at":1789420632.5424953,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450856","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-a07b8d55d9-a1","created_at":1789420632.6082149,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450857","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-143b8a233e-a1","created_at":1789420632.648734,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450858","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-f1b7b36d2a-a1","created_at":1789420632.7246292,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-f1b7b36d2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450859","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-d183a88991-a1","created_at":1789420632.7618475,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-d183a88991-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450860","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-882cb55e6f-a1","created_at":1789420632.8099668,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-882cb55e6f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450861","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-1ec3ab685a-a1","created_at":1789420632.8441193,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-1ec3ab685a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450862","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-0d25249e11-a1","created_at":1789420632.918382,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-0d25249e11-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450863","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-0cbc182214-a1","created_at":1789420632.9575224,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-0cbc182214-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450864","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","job_name":"wsd-dad07e1e-h20-ee74b9df2a-a1","created_at":1789412956.285909,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-dad07e1e-h20-ee74b9df2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/request.json"],"submission_uncertain":false,"scheduler":{"state":"FAILED","exit_code":"2:0"},"job_id":"2438299","error":"job 2438299 ended {'state': 'FAILED', 'exit_code': '2:0'}","campaign_identity":"dad07e1e4aa97cce6a325acd3d42c55fb5dbf52afbac97da9032eeb2832522a7"},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":2,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-002","job_name":"wsd-dee9b281-h20-ee74b9df2a-a2","created_at":1789418256.2847333,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-dee9b281-h20-ee74b9df2a-a2","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-002/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-002/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"FAILED","exit_code":"1:0"},"job_id":"2445945","error":"job 2445945 ended {'state': 'FAILED', 'exit_code': '1:0'}","campaign_identity":"dee9b2814cd7adeada424f28618dc84f109a6d41eee5c5ef1d2a8c6cd5560454"},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":3,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003","job_name":"wsd-5cc9d7b3-h20-ee74b9df2a-a3","created_at":1789420026.7726045,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-ee74b9df2a-a3","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2449510","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-29e6b737ab-a1","created_at":1789420633.0388203,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-29e6b737ab-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450865","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-4830aca569-a1","created_at":1789420633.0861735,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4830aca569-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450866","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-0fc272fc70-a1","created_at":1789420633.1531787,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450867","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-430ecc523e-a1","created_at":1789420633.2737474,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450868","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-d3f3cbdf31-a1","created_at":1789420633.4054582,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450869","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-520b13334b-a1","created_at":1789420633.4749317,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450870","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-95c4a6ab18-a1","created_at":1789420633.5695305,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450871","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-4ca8e83fae-a1","created_at":1789420633.632702,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450872","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-9b47563530-a1","created_at":1789420633.7113757,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-9b47563530-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450873","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-5c6c9d7536-a1","created_at":1789420633.7736192,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-5c6c9d7536-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450874","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-0118366529-a1","created_at":1789420633.8133297,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-0118366529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450875","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-37891cbc6f-a1","created_at":1789420633.8672874,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-37891cbc6f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450876","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-48600ea68b-a1","created_at":1789420633.931995,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-48600ea68b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450877","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-db42176d27-a1","created_at":1789420633.9958823,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-db42176d27-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450878","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-dc8efb2da0-a1","created_at":1789420634.0749042,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-dc8efb2da0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450879","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-bb9e12b405-a1","created_at":1789420634.1488664,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-bb9e12b405-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450880","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-849c783c12-a1","created_at":1789420634.202709,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-849c783c12-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450881","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-eec45e0f9f-a1","created_at":1789420634.276063,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450882","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-6d5cc9c704-a1","created_at":1789420634.3297784,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450883","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-7aa2c7ed9e-a1","created_at":1789420634.4235246,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450884","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-56405656d1-a1","created_at":1789420634.5521884,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450885","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-41875d52de-a1","created_at":1789420634.6011944,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450886","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-175a2e1097-a1","created_at":1789420634.693243,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450887","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-e601abd726-a1","created_at":1789420634.7681532,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-e601abd726-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450888","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-4e92a9610d-a1","created_at":1789420634.8599186,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4e92a9610d-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450889","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-e6d2b88d38-a1","created_at":1789420635.0168624,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-e6d2b88d38-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450890","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-ebd3071679-a1","created_at":1789420635.1298933,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-ebd3071679-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2450891","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-81a748b4a6-a1","created_at":1789421215.8349173,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-81a748b4a6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452072","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-e31c3b45a3-a1","created_at":1789421397.5002673,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-e31c3b45a3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452420","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-46824fbaa2-a1","created_at":1789421397.6271534,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-46824fbaa2-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452421","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-39417d41be-a1","created_at":1789421397.7601476,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-39417d41be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452422","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-4b87176d52-a1","created_at":1789421397.861317,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4b87176d52-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452423","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-dcd72cf962-a1","created_at":1789421397.9756508,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452424","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-475d380584-a1","created_at":1789421398.1278486,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452426","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-4b3c68558c-a1","created_at":1789421398.2574317,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452427","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-094dc7b901-a1","created_at":1789421398.3369708,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452428","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-4311bc41b1-a1","created_at":1789421398.5202525,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452429","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-6686968eba-a1","created_at":1789421398.6120656,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452430","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-c70b7bf720-a1","created_at":1789421398.793663,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-c70b7bf720-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452431","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-4c212235b4-a1","created_at":1789421398.870746,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-4c212235b4-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452432","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-13904a8ece-a1","created_at":1789421399.1037235,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-13904a8ece-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452433","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-a372cd5cdf-a1","created_at":1789421399.3345752,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-a372cd5cdf-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452434","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-bdd6c43135-a1","created_at":1789421399.6563225,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-bdd6c43135-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452439","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-60ee25a57b-a1","created_at":1789421399.8534014,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-60ee25a57b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"FAILED","exit_code":"1:0"},"job_id":"2452441","error":"job 2452441 ended {'state': 'FAILED', 'exit_code': '1:0'}"},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s44","number":2,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002","job_name":"wsd-5cc9d7b3-h20-60ee25a57b-a2","created_at":1789422479.062039,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-60ee25a57b-a2","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2454799","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-3b14b48bcd-a1","created_at":1789421399.9625385,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-3b14b48bcd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452443","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-b6a6d75078-a1","created_at":1789421400.0630739,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-b6a6d75078-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452447","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-95a138a529-a1","created_at":1789421400.180844,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-95a138a529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452449","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-ee9d1f4938-a1","created_at":1789421400.3588994,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452450","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-194749bb1c-a1","created_at":1789421400.447093,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452451","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-d42e2380d6-a1","created_at":1789421400.6370702,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452452","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-60c17463db-a1","created_at":1789421400.787323,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452453","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-b96e253615-a1","created_at":1789421400.8718545,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452454","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-be38278a7c-a1","created_at":1789421400.9812982,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452455","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-86a88f87be-a1","created_at":1789421401.2410114,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-86a88f87be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452456","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-e1bd62be9c-a1","created_at":1789421401.4083312,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-e1bd62be9c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452457","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-95f05450ca-a1","created_at":1789421401.651982,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-95f05450ca-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452458","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-019b523f21-a1","created_at":1789421401.938353,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-019b523f21-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452459","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-0321b9baa6-a1","created_at":1789421402.040926,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-0321b9baa6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452460","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-2240587c00-a1","created_at":1789421402.4956064,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-2240587c00-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452461","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-49f8a4199f-a1","created_at":1789421402.7280226,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-49f8a4199f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452462","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-c2bd3cb573-a1","created_at":1789421403.002416,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-c2bd3cb573-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452463","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-f8bdb01989-a1","created_at":1789421403.2430239,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-f8bdb01989-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452465","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-a8c1519773-a1","created_at":1789421403.5716994,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452466","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-0aed30c2b3-a1","created_at":1789421403.7177985,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452467","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-9c249fbf81-a1","created_at":1789421403.8269358,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452468","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h20-7fa76e51b0-a1","created_at":1789421404.0622091,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452469","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h20-8c3c99025e-a1","created_at":1789421404.3291528,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"FAILED","exit_code":"1:0"},"job_id":"2452470","error":"job 2452470 ended {'state': 'FAILED', 'exit_code': '1:0'}"},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":2,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002","job_name":"wsd-5cc9d7b3-h20-8c3c99025e-a2","created_at":1789422479.2990746,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-8c3c99025e-a2","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2454800","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h20-ffb2bac250-a1","created_at":1789421404.5086966,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h20-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2452471","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-367f579315-a1","created_at":1789423157.0651717,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-367f579315-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456190","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-b95fab87d6-a1","created_at":1789423157.7964807,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-b95fab87d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456193","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-e86c77c879-a1","created_at":1789423157.9359422,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-e86c77c879-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456200","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-8a5a73b5f9-a1","created_at":1789423158.003045,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456205","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-5f9d870038-a1","created_at":1789423158.1744301,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456206","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-66563544e3-a1","created_at":1789423158.2435155,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456207","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-14813e27f8-a1","created_at":1789423158.4965858,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456208","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-a07b8d55d9-a1","created_at":1789423158.6871738,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456209","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-143b8a233e-a1","created_at":1789423158.8295836,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456210","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-ee74b9df2a-a1","created_at":1789423158.9271955,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-ee74b9df2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456211","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-29e6b737ab-a1","created_at":1789423159.030313,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-29e6b737ab-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456213","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-4830aca569-a1","created_at":1789423159.1340673,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-4830aca569-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456214","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-0fc272fc70-a1","created_at":1789423159.2694094,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456215","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-430ecc523e-a1","created_at":1789423159.396973,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456216","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-d3f3cbdf31-a1","created_at":1789423159.5463636,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456217","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-520b13334b-a1","created_at":1789423159.6402676,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456218","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-95c4a6ab18-a1","created_at":1789423159.77656,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456219","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-4ca8e83fae-a1","created_at":1789423159.8726676,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456220","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-dc8efb2da0-a1","created_at":1789423159.9924397,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-dc8efb2da0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456221","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-bb9e12b405-a1","created_at":1789423160.3505497,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-bb9e12b405-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456269","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-849c783c12-a1","created_at":1789423162.2682939,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-849c783c12-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456270","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-eec45e0f9f-a1","created_at":1789423162.3857248,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456271","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-6d5cc9c704-a1","created_at":1789423162.5197191,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456272","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-7aa2c7ed9e-a1","created_at":1789423162.6198797,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456273","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-56405656d1-a1","created_at":1789423162.7201188,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456274","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-41875d52de-a1","created_at":1789423162.8034837,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456275","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-175a2e1097-a1","created_at":1789423162.9030194,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456276","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-46824fbaa2-a1","created_at":1789423162.9938915,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-46824fbaa2-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456277","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-39417d41be-a1","created_at":1789423163.1025174,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-39417d41be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456278","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-4b87176d52-a1","created_at":1789423163.2064357,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-4b87176d52-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456279","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-dcd72cf962-a1","created_at":1789423163.2917156,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456280","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-475d380584-a1","created_at":1789423163.3687782,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456281","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-4b3c68558c-a1","created_at":1789423163.4335413,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456282","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-094dc7b901-a1","created_at":1789423163.5008497,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456283","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-4311bc41b1-a1","created_at":1789423163.611973,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456284","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-6686968eba-a1","created_at":1789423163.681364,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456285","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-3b14b48bcd-a1","created_at":1789423163.7562575,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-3b14b48bcd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456286","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-b6a6d75078-a1","created_at":1789423163.8534899,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-b6a6d75078-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456287","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-95a138a529-a1","created_at":1789423163.9301436,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-95a138a529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456288","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-ee9d1f4938-a1","created_at":1789423164.040787,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456289","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-194749bb1c-a1","created_at":1789423164.1594806,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456290","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-d42e2380d6-a1","created_at":1789423164.2857904,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456291","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-60c17463db-a1","created_at":1789423164.37814,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456292","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-b96e253615-a1","created_at":1789423164.4764316,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456293","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-be38278a7c-a1","created_at":1789423164.5738778,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456294","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-49f8a4199f-a1","created_at":1789423164.7753232,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-49f8a4199f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456295","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-c2bd3cb573-a1","created_at":1789423164.9021046,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-c2bd3cb573-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456296","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-f8bdb01989-a1","created_at":1789423165.0341327,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-f8bdb01989-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2456297","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-a8c1519773-a1","created_at":1789423665.4882863,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2457298","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-0aed30c2b3-a1","created_at":1789423665.6098578,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2457299","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-9c249fbf81-a1","created_at":1789423665.6946213,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2457300","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h40-7fa76e51b0-a1","created_at":1789423665.783218,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2457301","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h40-8c3c99025e-a1","created_at":1789423665.916419,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2457302","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h40-ffb2bac250-a1","created_at":1789423666.0106902,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h40-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2457303","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-367f579315-a1","created_at":1789424467.9682267,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-367f579315-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458992","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-b95fab87d6-a1","created_at":1789424468.0724297,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-b95fab87d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458993","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-e86c77c879-a1","created_at":1789424468.227293,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-e86c77c879-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458994","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-8a5a73b5f9-a1","created_at":1789424468.3223622,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458995","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-5f9d870038-a1","created_at":1789424468.4582856,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458996","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-66563544e3-a1","created_at":1789424468.729649,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458997","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-14813e27f8-a1","created_at":1789424468.8256114,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458998","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-a07b8d55d9-a1","created_at":1789424468.9296865,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2458999","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-143b8a233e-a1","created_at":1789424469.02536,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459000","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-ee74b9df2a-a1","created_at":1789424469.2047782,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-ee74b9df2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459001","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-29e6b737ab-a1","created_at":1789424469.255579,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-29e6b737ab-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459002","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-4830aca569-a1","created_at":1789424469.4194477,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-4830aca569-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459003","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-0fc272fc70-a1","created_at":1789424469.513424,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459004","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-430ecc523e-a1","created_at":1789424469.6707518,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459005","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-d3f3cbdf31-a1","created_at":1789424469.7518938,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459006","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-520b13334b-a1","created_at":1789424470.0064497,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459007","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-95c4a6ab18-a1","created_at":1789424470.1850808,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459008","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-4ca8e83fae-a1","created_at":1789424470.2602732,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459009","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-dc8efb2da0-a1","created_at":1789424470.4944184,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-dc8efb2da0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459010","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-bb9e12b405-a1","created_at":1789424470.6379318,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-bb9e12b405-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459011","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-849c783c12-a1","created_at":1789424470.71681,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-849c783c12-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459012","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-eec45e0f9f-a1","created_at":1789424470.9108818,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459013","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-6d5cc9c704-a1","created_at":1789424470.9638674,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459017","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-7aa2c7ed9e-a1","created_at":1789424471.4374654,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459020","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-56405656d1-a1","created_at":1789424471.6845589,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459027","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-41875d52de-a1","created_at":1789424471.851588,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459030","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-175a2e1097-a1","created_at":1789424472.1548553,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459035","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-46824fbaa2-a1","created_at":1789424472.5480862,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-46824fbaa2-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459036","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-39417d41be-a1","created_at":1789424472.719637,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-39417d41be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459037","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-4b87176d52-a1","created_at":1789424472.906604,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-4b87176d52-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459038","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-dcd72cf962-a1","created_at":1789424473.119912,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459039","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-475d380584-a1","created_at":1789424473.222577,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459040","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-4b3c68558c-a1","created_at":1789424473.381165,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459059","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-094dc7b901-a1","created_at":1789424481.5171185,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459060","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-4311bc41b1-a1","created_at":1789424481.74158,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459061","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-6686968eba-a1","created_at":1789424482.053067,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459062","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-3b14b48bcd-a1","created_at":1789424482.1061952,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-3b14b48bcd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459063","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-b6a6d75078-a1","created_at":1789424482.395302,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-b6a6d75078-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459064","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-95a138a529-a1","created_at":1789424482.502611,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-95a138a529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459065","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-ee9d1f4938-a1","created_at":1789424482.6938927,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459066","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-194749bb1c-a1","created_at":1789424482.774378,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459067","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-d42e2380d6-a1","created_at":1789424482.9121666,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459068","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-60c17463db-a1","created_at":1789424483.1558726,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459069","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-b96e253615-a1","created_at":1789424483.2349038,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459070","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-be38278a7c-a1","created_at":1789424483.3666003,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459071","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-49f8a4199f-a1","created_at":1789424483.4353716,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-49f8a4199f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459072","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-c2bd3cb573-a1","created_at":1789424483.5049074,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-c2bd3cb573-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459073","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-f8bdb01989-a1","created_at":1789424483.5705745,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-f8bdb01989-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2459074","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-a8c1519773-a1","created_at":1789425246.5791733,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2460653","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-0aed30c2b3-a1","created_at":1789425246.6799743,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2460654","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-9c249fbf81-a1","created_at":1789425298.6057844,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2460783","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h80-7fa76e51b0-a1","created_at":1789425298.697634,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2460784","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h80-8c3c99025e-a1","created_at":1789425298.7669566,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2460785","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h80-ffb2bac250-a1","created_at":1789425298.8691945,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h80-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2460786","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-8a5a73b5f9-a1","created_at":1789426186.4514565,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461519","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-5f9d870038-a1","created_at":1789426188.7692995,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461523","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-66563544e3-a1","created_at":1789426188.841113,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461524","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-14813e27f8-a1","created_at":1789426188.9479501,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461526","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-a07b8d55d9-a1","created_at":1789426189.6413019,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461527","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-143b8a233e-a1","created_at":1789426189.7434452,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461528","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-0fc272fc70-a1","created_at":1789426189.8502102,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461529","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-430ecc523e-a1","created_at":1789426189.9256258,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461548","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-d3f3cbdf31-a1","created_at":1789426190.0153136,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461549","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-520b13334b-a1","created_at":1789426190.163671,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461550","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-95c4a6ab18-a1","created_at":1789426190.238572,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461551","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-4ca8e83fae-a1","created_at":1789426190.3303328,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461552","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-eec45e0f9f-a1","created_at":1789426190.447748,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461553","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-6d5cc9c704-a1","created_at":1789426190.5899403,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461554","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-7aa2c7ed9e-a1","created_at":1789426190.7116034,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461555","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-56405656d1-a1","created_at":1789426190.8186305,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461556","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-41875d52de-a1","created_at":1789426190.948457,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461557","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-175a2e1097-a1","created_at":1789426191.2182393,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461558","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-dcd72cf962-a1","created_at":1789426191.3491912,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461559","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-475d380584-a1","created_at":1789426191.4516728,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461560","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-4b3c68558c-a1","created_at":1789426191.572069,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461561","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-094dc7b901-a1","created_at":1789426191.6595914,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461562","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-4311bc41b1-a1","created_at":1789426191.766157,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461563","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-6686968eba-a1","created_at":1789426191.873722,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461564","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-ee9d1f4938-a1","created_at":1789426191.9623208,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461565","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-194749bb1c-a1","created_at":1789426192.0692582,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461566","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-d42e2380d6-a1","created_at":1789426192.1691232,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461567","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-60c17463db-a1","created_at":1789426192.3028302,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461568","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-b96e253615-a1","created_at":1789426192.4234962,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461569","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-be38278a7c-a1","created_at":1789426192.4894285,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461570","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-a8c1519773-a1","created_at":1789426192.5959814,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461571","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-0aed30c2b3-a1","created_at":1789426192.6987913,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461572","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-9c249fbf81-a1","created_at":1789426192.9204133,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461573","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h120-7fa76e51b0-a1","created_at":1789426192.9913595,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461574","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h120-8c3c99025e-a1","created_at":1789426193.128181,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461575","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h120-ffb2bac250-a1","created_at":1789426193.2636616,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h120-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2461576","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-8a5a73b5f9-a1","created_at":1789427352.7635734,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462419","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-5f9d870038-a1","created_at":1789427352.9432094,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462420","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-66563544e3-a1","created_at":1789427353.369086,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462421","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-14813e27f8-a1","created_at":1789427353.5423763,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462422","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-a07b8d55d9-a1","created_at":1789427353.6627202,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462423","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-143b8a233e-a1","created_at":1789427353.9595735,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462424","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-0fc272fc70-a1","created_at":1789427354.091337,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462425","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-430ecc523e-a1","created_at":1789427354.3656533,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462426","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-d3f3cbdf31-a1","created_at":1789427354.5161676,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462427","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-520b13334b-a1","created_at":1789427354.683458,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462428","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-95c4a6ab18-a1","created_at":1789427354.8264277,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462429","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-4ca8e83fae-a1","created_at":1789427354.982332,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462430","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-eec45e0f9f-a1","created_at":1789427355.138378,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462431","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-6d5cc9c704-a1","created_at":1789427355.2530618,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462432","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-7aa2c7ed9e-a1","created_at":1789427355.406531,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462433","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-56405656d1-a1","created_at":1789427355.5504787,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462434","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-41875d52de-a1","created_at":1789427355.6972456,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462436","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-175a2e1097-a1","created_at":1789427355.7993004,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462437","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-dcd72cf962-a1","created_at":1789427356.0012553,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462438","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-475d380584-a1","created_at":1789427356.320935,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462439","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-4b3c68558c-a1","created_at":1789427356.51479,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462440","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-094dc7b901-a1","created_at":1789427356.603896,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462441","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-4311bc41b1-a1","created_at":1789427356.6861331,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462442","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-6686968eba-a1","created_at":1789427356.8193438,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462443","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-ee9d1f4938-a1","created_at":1789427356.9232328,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462444","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-194749bb1c-a1","created_at":1789427357.0717258,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462445","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-d42e2380d6-a1","created_at":1789427357.3167503,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462446","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-60c17463db-a1","created_at":1789427357.4639034,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462447","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-b96e253615-a1","created_at":1789427357.5878973,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462448","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-be38278a7c-a1","created_at":1789427357.6908925,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462449","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-a8c1519773-a1","created_at":1789427357.8149633,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462450","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-0aed30c2b3-a1","created_at":1789427358.0149076,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462451","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-9c249fbf81-a1","created_at":1789427358.1757288,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462452","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"wsd-5cc9d7b3-h160-7fa76e51b0-a1","created_at":1789427358.253781,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462453","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"wsd-5cc9d7b3-h160-8c3c99025e-a1","created_at":1789427358.384085,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462454","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"wsd-5cc9d7b3-h160-ffb2bac250-a1","created_at":1789427358.5162003,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=wsd-5cc9d7b3-h160-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2462455","validated":true}]},"awc8":{"identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","manifestSha256":"9014e37a32015802214e02323cafeb30331db0bfaa079610252e2c8eb0129961","manifestPath":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/campaign.json","definition":{"version":1,"base":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.0056,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"runs/packed8-20m-awc","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"grid":{"beta1":[0.9,0.95],"beta2":[0.9,0.98,0.999],"lr":[0.01,0.008,0.006,0.004,0.003,0.002],"seeds":[42,43,44]},"horizons":[{"horizon":20,"epochs":40,"tokens":408027136,"steps":3113,"checkpoint_epoch":35,"checkpoint_tokens":357040128,"stable_end":367224422.40000004,"retained_checkpoint_epoch":35},{"horizon":40,"epochs":80,"tokens":816185344,"steps":6227,"checkpoint_epoch":72,"checkpoint_tokens":734527488,"stable_end":734566809.6,"retained_checkpoint_epoch":72},{"horizon":80,"epochs":160,"tokens":1632239616,"steps":12453,"checkpoint_epoch":143,"checkpoint_tokens":1458831360,"stable_end":1469015654.4,"retained_checkpoint_epoch":143},{"horizon":120,"epochs":240,"tokens":2448424960,"steps":18680,"checkpoint_epoch":216,"checkpoint_tokens":2203582464,"stable_end":2203582464.0,"retained_checkpoint_epoch":216},{"horizon":160,"epochs":320,"tokens":3264610304,"steps":24907,"checkpoint_epoch":288,"checkpoint_tokens":2938109952,"stable_end":2938149273.6,"retained_checkpoint_epoch":null}],"code":{"scripts":{"common.py":"3aeb0d18e4e09ec68ace96b28c6f2650110613a3f5321b2a013f3a7de3efc6f7","sweep.py":"e01b17f0bf11d64ef03bbe432ecf292b6026fbade0f98b69ab7ebd8f842b4284","worker.py":"6a80606287cadc10ed250054223cb1fd7fb6c0e4c8b70a1359ed9a675aab1cb3","launcher.sh":"7fea0ed98f8acb51b5ef6a5c7dbff436407f586b61ac91c7b68f710644340f10","manage.sh":"195fe0806597a790be824a4e74d7478530ca6899620b4e0f9e06cdce22cea6da","environment.sh":"380a7ca53bda84b7833f3857160fd7d511c61f611a4b1ff98f5a3bf964867e44"},"package":"0ef51b878b33c5375aa24d9c965feb590cfaa3411e1e62d32d15b0bb386909c1","dependencies":{"pyproject.toml":"6194efd1e3b255670aa8a9ee57d9585a5aa5207ff68fe81c4d38615322c6309f","uv.lock":"11fe47e1e133585bfc4fc83fd2498320dfcb349655e565a166489c668447cb6f"},"build_inputs":{"README.md":"5049d9bfca2069db79a6371a8e026d9ef4fea8d9061a644220e0138477e42bb4",".python-version":"7b55f8e67b5623c4bef3fa691288da9437d79d3aba156de48d481db32ac7d16d"},"recipes":{"20m.yaml":"19739a9bb22f20b2d3052d3df353156c29b42272e63f3d83e2b8b9bb52bfe880","wsd.yaml":"b995858ae6ef271bef84c972e7eb2bce0cecb45b177a97fddf22809a251d99c2","packed8-20m-awc.yaml":"1cfa79faad76477b0fda63e8a8ca0de0b1825388aa8e16f897aae5d81b62c4f6"}},"cache_identity":"7618ae8749071f012cd3e3e27f96480ea18c5da5dca59cf90330f35beaa4da1c","validation_tokens":197411295,"max_concurrent":48,"walltime_hours":2,"account":"naiss2026-3-205-gpu","partition":"gpu","gpu":"nvidia_gh200_120gb:1","pilot":"b1-0.9-b2-0.98-lr-0.004-s42","selection":"mean-three-seed-full-validation","retention":"all-average-and-worker-final-weights-rolling-winner-states-needed-parent-bundles","evaluation_model":"arithmetic-parameter-average-of-eight-workers","horizon_accounting":"global-training-tokens-per-single-model-parameter-count","storage_guard_bytes":21474836480,"job_workspace_bytes":8589934592,"max_open_configurations":16},"completedAt":1789491342.2930546,"retention":[{"time":1789462868.1513531,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789464302.8932502,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789464414.246679,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt"]},{"time":1789464976.1746929,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789465179.9330325,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789465248.9239774,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789466398.2756262,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789466446.852361,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789466483.9302852,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789466604.0907438,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.safetensors"]},{"time":1789466864.090782,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-001/epoch-035.pt"]},{"time":1789467513.004485,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.safetensors"]},{"time":1789467873.9643874,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.safetensors"]},{"time":1789467940.6579025,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt"]},{"time":1789470615.7159243,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.safetensors"]},{"time":1789470739.4589603,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.safetensors"]},{"time":1789470956.7141652,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt"]},{"time":1789472120.4597223,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.safetensors"]},{"time":1789472560.659514,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.safetensors"]},{"time":1789472690.9390917,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt"]},{"time":1789473502.1124597,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.safetensors"]},{"time":1789473669.4396915,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.safetensors"]},{"time":1789473856.1733081,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt"]},{"time":1789474112.9144082,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-143.pt"]},{"time":1789475280.902064,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789475499.400403,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789476166.5703027,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789476243.1951644,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789476289.3649602,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt"]},{"time":1789476385.39739,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477030.0548508,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477148.2407818,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477202.387707,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477247.2194448,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477421.6713507,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477495.5863159,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477730.4292662,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477902.039017,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789477983.7411087,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789478218.846493,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789478451.9778762,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789478712.19463,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789478762.452258,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt"]},{"time":1789478807.5390346,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789479127.9392383,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789479353.966115,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789479440.9119074,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789479660.832078,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789479858.668338,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.safetensors"]},{"time":1789479947.4442327,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.safetensors","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt"]},{"time":1789481364.4546056,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt"]},{"time":1789481408.8303509,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt"]},{"time":1789481516.5330257,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt"]},{"time":1789481746.5723372,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt"]},{"time":1789481969.309445,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt"]},{"time":1789482037.9889038,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt"]},{"time":1789482944.094995,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt"]},{"time":1789482996.0482304,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt"]},{"time":1789483042.1951392,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt"]},{"time":1789483085.46217,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt"]},{"time":1789484156.4642196,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt"]},{"time":1789484192.8426046,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt"]},{"time":1789484229.6149614,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt"]},{"time":1789484420.4355392,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt"]},{"time":1789485129.933311,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt"]},{"time":1789485198.670299,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt"]},{"time":1789485343.4352434,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt"]},{"time":1789485381.5586994,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt"]},{"time":1789486071.412673,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt"]},{"time":1789486230.5025191,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt"]},{"time":1789486268.2653067,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt"]},{"time":1789486366.5709567,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt"]},{"time":1789486494.42719,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt"]},{"time":1789486566.4831684,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt"]},{"time":1789491303.0104985,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt"]},{"time":1789491339.9861658,"removed":["/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt"]}],"attempts":[{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.01-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001","job_name":"awc8-3487e1c9-h20-4eea11b8da-a1","created_at":1789462868.4840024,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4eea11b8da-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478551","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.01-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001","job_name":"awc8-3487e1c9-h20-2a2e750fed-a1","created_at":1789462868.5492525,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-2a2e750fed-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478552","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.01-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001","job_name":"awc8-3487e1c9-h20-d8b0b7c8f8-a1","created_at":1789462868.598637,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-d8b0b7c8f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478553","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001","job_name":"awc8-3487e1c9-h20-f415dc1dcc-a1","created_at":1789462868.6418018,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-f415dc1dcc-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478554","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001","job_name":"awc8-3487e1c9-h20-498332c166-a1","created_at":1789462868.7883112,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-498332c166-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478555","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001","job_name":"awc8-3487e1c9-h20-31e6b398d5-a1","created_at":1789462868.8693123,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-31e6b398d5-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478556","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h20-ac5caa08db-a1","created_at":1789462869.0583816,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-ac5caa08db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478557","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h20-46c6775a30-a1","created_at":1789462869.1156383,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-46c6775a30-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478558","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h20-7d4714319e-a1","created_at":1789462869.169299,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-7d4714319e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478559","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h20-367f579315-a1","created_at":1789462869.5272884,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-367f579315-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478560","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h20-b95fab87d6-a1","created_at":1789462869.6383328,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-b95fab87d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478561","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h20-e86c77c879-a1","created_at":1789462869.849295,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-e86c77c879-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478563","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h20-8a5a73b5f9-a1","created_at":1789462870.0597112,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478564","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h20-5f9d870038-a1","created_at":1789462870.1364572,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478565","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h20-66563544e3-a1","created_at":1789462870.2973683,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478566","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h20-14813e27f8-a1","created_at":1789462870.381873,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478567","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h20-a07b8d55d9-a1","created_at":1789462870.5280962,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478568","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h20-143b8a233e-a1","created_at":1789462870.5817347,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478569","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.01-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001","job_name":"awc8-3487e1c9-h20-c86670a3a0-a1","created_at":1789462870.645579,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-c86670a3a0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478570","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.01-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001","job_name":"awc8-3487e1c9-h20-0ff8c368c3-a1","created_at":1789462870.8283062,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0ff8c368c3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478571","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.01-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001","job_name":"awc8-3487e1c9-h20-70793c324b-a1","created_at":1789462870.900912,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-70793c324b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478572","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001","job_name":"awc8-3487e1c9-h20-f1b7b36d2a-a1","created_at":1789462871.0426962,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-f1b7b36d2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478573","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001","job_name":"awc8-3487e1c9-h20-d183a88991-a1","created_at":1789462871.1112416,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-d183a88991-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478574","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001","job_name":"awc8-3487e1c9-h20-882cb55e6f-a1","created_at":1789462871.1730058,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-882cb55e6f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478575","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h20-1ec3ab685a-a1","created_at":1789462871.310769,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-1ec3ab685a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478576","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h20-0d25249e11-a1","created_at":1789462871.39602,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0d25249e11-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478577","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h20-0cbc182214-a1","created_at":1789462871.581659,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0cbc182214-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478578","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h20-ee74b9df2a-a1","created_at":1789460673.4987564,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-ee74b9df2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2477538","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h20-29e6b737ab-a1","created_at":1789462871.643671,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-29e6b737ab-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478579","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h20-4830aca569-a1","created_at":1789462871.7973988,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4830aca569-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478580","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h20-0fc272fc70-a1","created_at":1789462871.8905187,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478581","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h20-430ecc523e-a1","created_at":1789462872.0976832,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478582","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h20-d3f3cbdf31-a1","created_at":1789462872.3620772,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478583","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h20-520b13334b-a1","created_at":1789462872.4741175,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478584","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h20-95c4a6ab18-a1","created_at":1789462872.64424,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478585","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h20-4ca8e83fae-a1","created_at":1789462872.8450766,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478586","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.01-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001","job_name":"awc8-3487e1c9-h20-0cb771d7d4-a1","created_at":1789462872.9523242,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0cb771d7d4-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478587","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.01-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001","job_name":"awc8-3487e1c9-h20-27ad5a42cd-a1","created_at":1789462873.1447935,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-27ad5a42cd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478588","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.01-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001","job_name":"awc8-3487e1c9-h20-65d5224e84-a1","created_at":1789462873.3459375,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-65d5224e84-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478589","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001","job_name":"awc8-3487e1c9-h20-9b47563530-a1","created_at":1789462873.432433,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-9b47563530-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478590","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001","job_name":"awc8-3487e1c9-h20-5c6c9d7536-a1","created_at":1789462873.610372,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-5c6c9d7536-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478591","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001","job_name":"awc8-3487e1c9-h20-0118366529-a1","created_at":1789462873.7021296,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0118366529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478592","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h20-37891cbc6f-a1","created_at":1789462873.840576,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-37891cbc6f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478593","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h20-48600ea68b-a1","created_at":1789462873.9318142,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-48600ea68b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478594","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h20-db42176d27-a1","created_at":1789462874.1200829,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-db42176d27-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478595","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h20-dc8efb2da0-a1","created_at":1789462881.1167831,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-dc8efb2da0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478596","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h20-bb9e12b405-a1","created_at":1789462881.2804918,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-bb9e12b405-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478597","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h20-849c783c12-a1","created_at":1789462881.3368087,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-849c783c12-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2478598","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h20-eec45e0f9f-a1","created_at":1789464303.3619075,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479238","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h20-6d5cc9c704-a1","created_at":1789464303.5521019,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479239","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h20-7aa2c7ed9e-a1","created_at":1789464303.6789792,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479240","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h20-56405656d1-a1","created_at":1789464303.8278112,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479241","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h20-41875d52de-a1","created_at":1789464303.9660647,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479242","validated":true},{"horizon":20,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h20-175a2e1097-a1","created_at":1789464304.1140718,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479243","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.01-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001","job_name":"awc8-3487e1c9-h20-7e34f0218b-a1","created_at":1789464304.233713,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-7e34f0218b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479244","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.01-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001","job_name":"awc8-3487e1c9-h20-07b80606bf-a1","created_at":1789464304.3664877,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-07b80606bf-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479245","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.01-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001","job_name":"awc8-3487e1c9-h20-941f10a65a-a1","created_at":1789464304.559237,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-941f10a65a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479246","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001","job_name":"awc8-3487e1c9-h20-e601abd726-a1","created_at":1789464304.7001438,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-e601abd726-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479247","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001","job_name":"awc8-3487e1c9-h20-4e92a9610d-a1","created_at":1789464304.8206022,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4e92a9610d-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479248","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001","job_name":"awc8-3487e1c9-h20-e6d2b88d38-a1","created_at":1789464304.9632492,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-e6d2b88d38-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479249","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h20-ebd3071679-a1","created_at":1789464305.1106424,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-ebd3071679-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479250","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h20-81a748b4a6-a1","created_at":1789464305.1948771,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-81a748b4a6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479251","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h20-e31c3b45a3-a1","created_at":1789464305.3149264,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-e31c3b45a3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479252","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h20-46824fbaa2-a1","created_at":1789464305.4193876,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-46824fbaa2-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479253","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h20-39417d41be-a1","created_at":1789464305.5751693,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-39417d41be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479254","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h20-4b87176d52-a1","created_at":1789464305.6956553,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4b87176d52-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479255","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h20-dcd72cf962-a1","created_at":1789464305.7933006,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479256","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h20-475d380584-a1","created_at":1789464305.8840005,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479257","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h20-4b3c68558c-a1","created_at":1789464305.9986844,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479259","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h20-094dc7b901-a1","created_at":1789464306.063324,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479260","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h20-4311bc41b1-a1","created_at":1789464306.2257938,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479261","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h20-6686968eba-a1","created_at":1789464306.2993283,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479262","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.01-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001","job_name":"awc8-3487e1c9-h20-f1914bd401-a1","created_at":1789464306.4115663,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-f1914bd401-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479263","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.01-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001","job_name":"awc8-3487e1c9-h20-61236acff8-a1","created_at":1789464306.5608807,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-61236acff8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479264","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.01-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001","job_name":"awc8-3487e1c9-h20-8bc2b61077-a1","created_at":1789464306.7574034,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-8bc2b61077-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479265","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001","job_name":"awc8-3487e1c9-h20-c70b7bf720-a1","created_at":1789464306.8410518,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-c70b7bf720-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479266","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001","job_name":"awc8-3487e1c9-h20-4c212235b4-a1","created_at":1789464306.9717877,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-4c212235b4-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479267","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001","job_name":"awc8-3487e1c9-h20-13904a8ece-a1","created_at":1789464307.1617537,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-13904a8ece-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479268","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h20-a372cd5cdf-a1","created_at":1789464307.2703552,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-a372cd5cdf-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479269","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h20-bdd6c43135-a1","created_at":1789464307.3898342,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-bdd6c43135-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479270","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h20-60ee25a57b-a1","created_at":1789464307.503085,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-60ee25a57b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479271","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h20-3b14b48bcd-a1","created_at":1789464307.5866704,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-3b14b48bcd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479272","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h20-b6a6d75078-a1","created_at":1789464307.6880162,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-b6a6d75078-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479273","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h20-95a138a529-a1","created_at":1789464307.8328865,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-95a138a529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479274","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h20-ee9d1f4938-a1","created_at":1789464414.5599322,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479308","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h20-194749bb1c-a1","created_at":1789464414.7496676,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479309","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h20-d42e2380d6-a1","created_at":1789464414.8830216,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479310","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h20-60c17463db-a1","created_at":1789464415.0641968,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479311","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h20-b96e253615-a1","created_at":1789464415.2313035,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479312","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h20-be38278a7c-a1","created_at":1789464415.3763955,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479313","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.01-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001","job_name":"awc8-3487e1c9-h20-85499f677d-a1","created_at":1789464415.522908,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-85499f677d-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479314","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.01-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001","job_name":"awc8-3487e1c9-h20-49bd5545be-a1","created_at":1789464415.6444232,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-49bd5545be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479315","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.01-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001","job_name":"awc8-3487e1c9-h20-ee166fa938-a1","created_at":1789464415.7818077,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-ee166fa938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479316","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.008-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001","job_name":"awc8-3487e1c9-h20-86a88f87be-a1","created_at":1789464415.9238853,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-86a88f87be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479317","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.008-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001","job_name":"awc8-3487e1c9-h20-e1bd62be9c-a1","created_at":1789464416.0636032,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-e1bd62be9c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479318","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.008-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001","job_name":"awc8-3487e1c9-h20-95f05450ca-a1","created_at":1789464416.2319584,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-95f05450ca-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479319","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h20-019b523f21-a1","created_at":1789464976.5966868,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-019b523f21-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479658","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h20-0321b9baa6-a1","created_at":1789464976.692054,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0321b9baa6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479659","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h20-2240587c00-a1","created_at":1789464976.789166,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-2240587c00-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479660","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h20-49f8a4199f-a1","created_at":1789464976.9216151,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-49f8a4199f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479661","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h20-c2bd3cb573-a1","created_at":1789464977.0153058,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-c2bd3cb573-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479662","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h20-f8bdb01989-a1","created_at":1789464977.148902,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-f8bdb01989-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479663","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h20-a8c1519773-a1","created_at":1789464977.2691305,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479664","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h20-0aed30c2b3-a1","created_at":1789464977.3779438,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479665","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h20-9c249fbf81-a1","created_at":1789464977.4420633,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479666","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h20-7fa76e51b0-a1","created_at":1789464977.5021176,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479667","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h20-8c3c99025e-a1","created_at":1789464977.628845,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479668","validated":true},{"horizon":20,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h20-ffb2bac250-a1","created_at":1789464977.7604327,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h20-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2479669","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h40-ac5caa08db-a1","created_at":1789466864.3770094,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-ac5caa08db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480625","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h40-46c6775a30-a1","created_at":1789466864.6697307,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-46c6775a30-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480626","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h40-7d4714319e-a1","created_at":1789466864.787094,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-7d4714319e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480627","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h40-367f579315-a1","created_at":1789466864.9189863,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-367f579315-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480628","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h40-b95fab87d6-a1","created_at":1789466865.071653,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-b95fab87d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480629","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h40-e86c77c879-a1","created_at":1789466865.2383037,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-e86c77c879-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480630","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h40-8a5a73b5f9-a1","created_at":1789466865.4106143,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480632","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h40-5f9d870038-a1","created_at":1789466867.0664532,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480633","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h40-66563544e3-a1","created_at":1789466867.2299047,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480634","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h40-14813e27f8-a1","created_at":1789466867.3919008,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480635","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h40-a07b8d55d9-a1","created_at":1789466867.515755,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480636","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h40-143b8a233e-a1","created_at":1789466867.641461,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480637","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h40-1ec3ab685a-a1","created_at":1789466867.8047795,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-1ec3ab685a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480638","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h40-0d25249e11-a1","created_at":1789466867.9404273,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-0d25249e11-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480639","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h40-0cbc182214-a1","created_at":1789466868.0631456,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-0cbc182214-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480640","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h40-ee74b9df2a-a1","created_at":1789466868.1691504,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-ee74b9df2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480641","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h40-29e6b737ab-a1","created_at":1789466868.3304331,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-29e6b737ab-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480642","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h40-4830aca569-a1","created_at":1789466868.4615626,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-4830aca569-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480643","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h40-0fc272fc70-a1","created_at":1789466868.6456914,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480644","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h40-430ecc523e-a1","created_at":1789466868.7600935,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480645","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h40-d3f3cbdf31-a1","created_at":1789466868.86026,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480646","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h40-520b13334b-a1","created_at":1789466869.0080736,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480647","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h40-95c4a6ab18-a1","created_at":1789466869.1744125,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480648","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h40-4ca8e83fae-a1","created_at":1789466869.3689644,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480649","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h40-37891cbc6f-a1","created_at":1789466869.512281,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-37891cbc6f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480650","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h40-48600ea68b-a1","created_at":1789466869.6718726,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-48600ea68b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480651","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h40-db42176d27-a1","created_at":1789466869.9270194,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-db42176d27-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480652","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h40-dc8efb2da0-a1","created_at":1789466870.0088375,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-dc8efb2da0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480653","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h40-bb9e12b405-a1","created_at":1789466870.1471033,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-bb9e12b405-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480654","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h40-849c783c12-a1","created_at":1789466870.370926,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-849c783c12-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480655","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h40-eec45e0f9f-a1","created_at":1789466870.5113585,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480656","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h40-6d5cc9c704-a1","created_at":1789466870.6741924,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480657","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h40-7aa2c7ed9e-a1","created_at":1789466870.7727315,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480658","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h40-56405656d1-a1","created_at":1789466870.867967,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480659","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h40-41875d52de-a1","created_at":1789466870.965635,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480660","validated":true},{"horizon":40,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h40-175a2e1097-a1","created_at":1789466871.152493,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480661","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h40-ebd3071679-a1","created_at":1789466871.316551,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-ebd3071679-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480662","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h40-81a748b4a6-a1","created_at":1789466871.4466383,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-81a748b4a6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480663","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h40-e31c3b45a3-a1","created_at":1789466871.5742464,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-e31c3b45a3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480664","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h40-46824fbaa2-a1","created_at":1789466871.6919458,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-46824fbaa2-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480665","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h40-39417d41be-a1","created_at":1789466871.8110452,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-39417d41be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480666","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h40-4b87176d52-a1","created_at":1789466871.9430013,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-4b87176d52-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480667","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h40-dcd72cf962-a1","created_at":1789466872.1238415,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480668","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h40-475d380584-a1","created_at":1789466872.2579694,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480669","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h40-4b3c68558c-a1","created_at":1789466872.435893,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480670","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h40-094dc7b901-a1","created_at":1789466872.5705028,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480671","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h40-4311bc41b1-a1","created_at":1789466872.719705,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480672","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h40-6686968eba-a1","created_at":1789466872.8569937,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2480673","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h40-a372cd5cdf-a1","created_at":1789467875.0431736,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-a372cd5cdf-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481085","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h40-bdd6c43135-a1","created_at":1789467875.2375212,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-bdd6c43135-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481086","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h40-60ee25a57b-a1","created_at":1789467875.4214933,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-60ee25a57b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481087","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h40-3b14b48bcd-a1","created_at":1789467875.58695,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-3b14b48bcd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481088","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h40-b6a6d75078-a1","created_at":1789467875.75232,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-b6a6d75078-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481089","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h40-95a138a529-a1","created_at":1789467875.9749305,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-95a138a529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481090","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h40-ee9d1f4938-a1","created_at":1789467876.1492026,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481091","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h40-194749bb1c-a1","created_at":1789467876.3035426,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481092","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h40-d42e2380d6-a1","created_at":1789467876.477168,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481093","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h40-60c17463db-a1","created_at":1789467876.7396786,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481094","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h40-b96e253615-a1","created_at":1789467876.8836288,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481095","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h40-be38278a7c-a1","created_at":1789467877.0223618,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481096","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h40-019b523f21-a1","created_at":1789467877.155451,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-019b523f21-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481097","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h40-0321b9baa6-a1","created_at":1789467877.3296232,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-0321b9baa6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481098","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h40-2240587c00-a1","created_at":1789467877.4876392,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-2240587c00-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481099","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h40-49f8a4199f-a1","created_at":1789467877.6651232,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-49f8a4199f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481100","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h40-c2bd3cb573-a1","created_at":1789467877.8170855,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-c2bd3cb573-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481101","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h40-f8bdb01989-a1","created_at":1789467877.9739006,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-f8bdb01989-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481102","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h40-a8c1519773-a1","created_at":1789467878.1682699,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481103","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h40-0aed30c2b3-a1","created_at":1789467878.3270288,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481104","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h40-9c249fbf81-a1","created_at":1789467878.4984088,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481105","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h40-7fa76e51b0-a1","created_at":1789467878.6728613,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481106","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h40-8c3c99025e-a1","created_at":1789467878.8803914,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481107","validated":true},{"horizon":40,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h40-ffb2bac250-a1","created_at":1789467879.0299728,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h40-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481108","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h80-ac5caa08db-a1","created_at":1789471193.8474865,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-ac5caa08db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481774","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h80-46c6775a30-a1","created_at":1789471194.1027107,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-46c6775a30-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481775","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h80-7d4714319e-a1","created_at":1789471194.3032844,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-7d4714319e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481776","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h80-367f579315-a1","created_at":1789471194.5881748,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-367f579315-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481777","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h80-b95fab87d6-a1","created_at":1789471194.8011172,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-b95fab87d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481778","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h80-e86c77c879-a1","created_at":1789471195.019171,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-e86c77c879-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481779","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h80-8a5a73b5f9-a1","created_at":1789471195.3101652,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481780","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h80-5f9d870038-a1","created_at":1789471195.5408854,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481781","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h80-66563544e3-a1","created_at":1789471195.7424557,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481782","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h80-14813e27f8-a1","created_at":1789471195.929011,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481783","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h80-a07b8d55d9-a1","created_at":1789471196.1199355,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481784","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h80-143b8a233e-a1","created_at":1789471196.325435,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481785","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h80-1ec3ab685a-a1","created_at":1789471196.5209913,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-1ec3ab685a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481786","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h80-0d25249e11-a1","created_at":1789471196.7368152,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-0d25249e11-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481787","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h80-0cbc182214-a1","created_at":1789471196.9735227,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-0cbc182214-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481788","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h80-ee74b9df2a-a1","created_at":1789471197.174813,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-ee74b9df2a-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481789","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h80-29e6b737ab-a1","created_at":1789471197.3825045,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-29e6b737ab-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481790","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h80-4830aca569-a1","created_at":1789471197.5848007,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-4830aca569-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481791","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h80-0fc272fc70-a1","created_at":1789471197.7937396,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481792","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h80-430ecc523e-a1","created_at":1789471198.0133996,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481793","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h80-d3f3cbdf31-a1","created_at":1789471198.260431,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481794","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h80-520b13334b-a1","created_at":1789471198.4036326,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481795","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h80-95c4a6ab18-a1","created_at":1789471198.6631174,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481796","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h80-4ca8e83fae-a1","created_at":1789471198.8251255,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481797","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h80-37891cbc6f-a1","created_at":1789471199.0550742,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-37891cbc6f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481798","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h80-48600ea68b-a1","created_at":1789471199.286284,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-48600ea68b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481799","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h80-db42176d27-a1","created_at":1789471199.4653668,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-db42176d27-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481800","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h80-dc8efb2da0-a1","created_at":1789471199.6745358,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-dc8efb2da0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481801","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h80-bb9e12b405-a1","created_at":1789471199.8563058,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-bb9e12b405-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481802","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h80-849c783c12-a1","created_at":1789471200.0588763,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-849c783c12-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481803","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h80-eec45e0f9f-a1","created_at":1789471200.301035,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481804","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h80-6d5cc9c704-a1","created_at":1789471200.515258,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481805","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h80-7aa2c7ed9e-a1","created_at":1789471200.7470598,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481806","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h80-56405656d1-a1","created_at":1789471200.953579,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481807","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h80-41875d52de-a1","created_at":1789471201.216974,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481808","validated":true},{"horizon":80,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h80-175a2e1097-a1","created_at":1789471201.4377751,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481809","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h80-ebd3071679-a1","created_at":1789471201.6569574,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-ebd3071679-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481810","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h80-81a748b4a6-a1","created_at":1789471201.8270037,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-81a748b4a6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481811","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h80-e31c3b45a3-a1","created_at":1789471202.0641057,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-e31c3b45a3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481812","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h80-46824fbaa2-a1","created_at":1789471202.2769127,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-46824fbaa2-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481813","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h80-39417d41be-a1","created_at":1789471202.5410295,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-39417d41be-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481814","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h80-4b87176d52-a1","created_at":1789471203.019537,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-4b87176d52-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481815","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h80-dcd72cf962-a1","created_at":1789471203.2769887,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481816","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h80-475d380584-a1","created_at":1789471203.44986,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481817","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h80-4b3c68558c-a1","created_at":1789471203.6472478,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481818","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h80-094dc7b901-a1","created_at":1789471203.8135483,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481819","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h80-4311bc41b1-a1","created_at":1789471203.9668202,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481820","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h80-6686968eba-a1","created_at":1789471204.161774,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2481821","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h80-a372cd5cdf-a1","created_at":1789472561.2225776,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-a372cd5cdf-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482106","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h80-bdd6c43135-a1","created_at":1789472561.4736638,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-bdd6c43135-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482107","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h80-60ee25a57b-a1","created_at":1789472561.6764743,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-60ee25a57b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482108","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h80-3b14b48bcd-a1","created_at":1789472561.8646634,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-3b14b48bcd-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482109","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h80-b6a6d75078-a1","created_at":1789472562.0813618,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-b6a6d75078-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482110","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h80-95a138a529-a1","created_at":1789472562.2569077,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-95a138a529-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482111","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h80-ee9d1f4938-a1","created_at":1789472562.4767096,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482112","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h80-194749bb1c-a1","created_at":1789472562.677945,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482113","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h80-d42e2380d6-a1","created_at":1789472563.4424086,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482114","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h80-60c17463db-a1","created_at":1789472563.6679819,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482115","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h80-b96e253615-a1","created_at":1789472563.9006104,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482116","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h80-be38278a7c-a1","created_at":1789472564.159487,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482117","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.006-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","job_name":"awc8-3487e1c9-h80-019b523f21-a1","created_at":1789472564.4039285,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-019b523f21-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482118","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.006-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","job_name":"awc8-3487e1c9-h80-0321b9baa6-a1","created_at":1789472564.6477983,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-0321b9baa6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482119","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.006-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","job_name":"awc8-3487e1c9-h80-2240587c00-a1","created_at":1789472564.8577518,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-2240587c00-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482120","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.004-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","job_name":"awc8-3487e1c9-h80-49f8a4199f-a1","created_at":1789472565.0738032,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-49f8a4199f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482121","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.004-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","job_name":"awc8-3487e1c9-h80-c2bd3cb573-a1","created_at":1789472565.2989154,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-c2bd3cb573-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482122","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.004-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","job_name":"awc8-3487e1c9-h80-f8bdb01989-a1","created_at":1789472565.5033607,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-f8bdb01989-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482123","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h80-a8c1519773-a1","created_at":1789472565.6889513,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482124","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h80-0aed30c2b3-a1","created_at":1789472565.8989577,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482125","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h80-9c249fbf81-a1","created_at":1789472566.093098,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482126","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h80-7fa76e51b0-a1","created_at":1789472566.3107264,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482127","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h80-8c3c99025e-a1","created_at":1789472566.4894366,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482128","validated":true},{"horizon":80,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h80-ffb2bac250-a1","created_at":1789472566.744065,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h80-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482129","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h120-8a5a73b5f9-a1","created_at":1789474113.9214797,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482589","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h120-5f9d870038-a1","created_at":1789474114.2961767,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482590","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h120-66563544e3-a1","created_at":1789474114.5763445,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482591","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h120-14813e27f8-a1","created_at":1789474115.0682635,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482592","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h120-a07b8d55d9-a1","created_at":1789474115.3599014,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482593","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h120-143b8a233e-a1","created_at":1789474115.6071372,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482594","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h120-0fc272fc70-a1","created_at":1789474115.7822893,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482595","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h120-430ecc523e-a1","created_at":1789474115.9744291,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482596","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h120-d3f3cbdf31-a1","created_at":1789474116.1653676,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482597","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h120-520b13334b-a1","created_at":1789474116.3756948,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482598","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h120-95c4a6ab18-a1","created_at":1789474116.592255,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482599","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h120-4ca8e83fae-a1","created_at":1789474116.7993515,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482600","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h120-eec45e0f9f-a1","created_at":1789474116.9703238,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482601","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h120-6d5cc9c704-a1","created_at":1789474117.1799293,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482602","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h120-7aa2c7ed9e-a1","created_at":1789474117.37884,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482603","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h120-56405656d1-a1","created_at":1789474117.5675073,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482604","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h120-41875d52de-a1","created_at":1789474117.7927074,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482605","validated":true},{"horizon":120,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h120-175a2e1097-a1","created_at":1789474118.0164254,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482606","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h120-dcd72cf962-a1","created_at":1789474118.2624712,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482607","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h120-475d380584-a1","created_at":1789474118.5180514,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482608","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h120-4b3c68558c-a1","created_at":1789474118.7450106,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482609","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h120-094dc7b901-a1","created_at":1789474118.9345622,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482610","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h120-4311bc41b1-a1","created_at":1789474119.1191735,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482611","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h120-6686968eba-a1","created_at":1789474119.3357656,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482612","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h120-ee9d1f4938-a1","created_at":1789474119.769025,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482613","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h120-194749bb1c-a1","created_at":1789474120.0680482,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482614","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h120-d42e2380d6-a1","created_at":1789474120.379508,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482615","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h120-60c17463db-a1","created_at":1789474120.7329724,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482617","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h120-b96e253615-a1","created_at":1789474121.1386364,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482618","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h120-be38278a7c-a1","created_at":1789474122.3568366,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482619","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h120-a8c1519773-a1","created_at":1789474123.0011046,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482620","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h120-0aed30c2b3-a1","created_at":1789474123.6299024,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482621","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h120-9c249fbf81-a1","created_at":1789474124.3412077,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482622","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h120-7fa76e51b0-a1","created_at":1789474125.080744,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482623","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h120-8c3c99025e-a1","created_at":1789474125.3451192,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482624","validated":true},{"horizon":120,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h120-ffb2bac250-a1","created_at":1789474125.6173885,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h120-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2482625","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h160-8a5a73b5f9-a1","created_at":1789480060.5868359,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-8a5a73b5f9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487090","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h160-5f9d870038-a1","created_at":1789480061.7949455,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-5f9d870038-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487091","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h160-66563544e3-a1","created_at":1789480062.0120118,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-66563544e3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487092","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h160-14813e27f8-a1","created_at":1789480064.3814938,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-14813e27f8-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487093","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h160-a07b8d55d9-a1","created_at":1789480064.6162753,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-a07b8d55d9-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487094","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h160-143b8a233e-a1","created_at":1789480066.7994556,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-143b8a233e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487095","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h160-0fc272fc70-a1","created_at":1789480067.028021,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-0fc272fc70-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487096","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h160-430ecc523e-a1","created_at":1789480069.2554796,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-430ecc523e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487099","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h160-d3f3cbdf31-a1","created_at":1789480069.48667,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-d3f3cbdf31-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487100","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h160-520b13334b-a1","created_at":1789480070.6961412,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-520b13334b-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487102","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h160-95c4a6ab18-a1","created_at":1789480070.9327142,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-95c4a6ab18-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487103","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h160-4ca8e83fae-a1","created_at":1789480071.1643248,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-4ca8e83fae-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487104","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h160-eec45e0f9f-a1","created_at":1789480071.4208446,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-eec45e0f9f-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487105","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h160-6d5cc9c704-a1","created_at":1789480071.6771986,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-6d5cc9c704-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487106","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h160-7aa2c7ed9e-a1","created_at":1789480071.9100835,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-7aa2c7ed9e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487107","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h160-56405656d1-a1","created_at":1789480072.1779482,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-56405656d1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487108","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h160-41875d52de-a1","created_at":1789480072.3967013,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-41875d52de-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487109","validated":true},{"horizon":160,"run":"b1-0.9-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h160-175a2e1097-a1","created_at":1789480072.6154013,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-175a2e1097-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487110","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h160-dcd72cf962-a1","created_at":1789480072.8499262,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-dcd72cf962-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487111","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h160-475d380584-a1","created_at":1789480073.1029952,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-475d380584-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487112","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h160-4b3c68558c-a1","created_at":1789480073.3687172,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-4b3c68558c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487113","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h160-094dc7b901-a1","created_at":1789480073.5724819,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-094dc7b901-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487114","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h160-4311bc41b1-a1","created_at":1789480073.8092325,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-4311bc41b1-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487115","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.9-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h160-6686968eba-a1","created_at":1789480074.0590236,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-6686968eba-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487116","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h160-ee9d1f4938-a1","created_at":1789480074.3240592,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-ee9d1f4938-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487117","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h160-194749bb1c-a1","created_at":1789480074.8149493,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-194749bb1c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487118","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h160-d42e2380d6-a1","created_at":1789480075.0484278,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-d42e2380d6-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487119","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h160-60c17463db-a1","created_at":1789480075.2924657,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-60c17463db-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487120","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h160-b96e253615-a1","created_at":1789480075.5956116,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-b96e253615-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487121","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.98-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h160-be38278a7c-a1","created_at":1789480075.7903445,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-be38278a7c-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487122","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.003-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","job_name":"awc8-3487e1c9-h160-a8c1519773-a1","created_at":1789480076.0052042,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-a8c1519773-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487123","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.003-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","job_name":"awc8-3487e1c9-h160-0aed30c2b3-a1","created_at":1789480076.2330265,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-0aed30c2b3-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487124","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.003-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","job_name":"awc8-3487e1c9-h160-9c249fbf81-a1","created_at":1789480076.460672,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-9c249fbf81-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487125","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.002-s42","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","job_name":"awc8-3487e1c9-h160-7fa76e51b0-a1","created_at":1789480076.6941857,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-7fa76e51b0-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487127","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.002-s43","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","job_name":"awc8-3487e1c9-h160-8c3c99025e-a1","created_at":1789480076.949742,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-8c3c99025e-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487128","validated":true},{"horizon":160,"run":"b1-0.95-b2-0.999-lr-0.002-s44","number":1,"directory":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","job_name":"awc8-3487e1c9-h160-ffb2bac250-a1","created_at":1789480077.186503,"command":["sbatch","--parsable","--account=naiss2026-3-205-gpu","--partition=gpu","--nodes=1","--ntasks=1","--gpus=nvidia_gh200_120gb:1","--time=02:00:00","--job-name=awc8-3487e1c9-h160-ffb2bac250-a1","--output=/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/slurm-%j.log","--export=ALL","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/launcher.sh","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/request.json","/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning"],"submission_uncertain":false,"scheduler":{"state":"COMPLETED","exit_code":"0:0"},"job_id":"2487129","validated":true}]}},"runs":{"sync-wsd-h20--0.008--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2450844","recipe_identity":"fc01b936f4e98df37aff5e4aeab1508053ac81ec703b37e41af6da964b9ae854","loss":3.577678713712776,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/result.json","sha256":"967cd7c35a58dbfe77e7b69c1b0de4941ba55ff2eb265d0f8d1912d27b2aafdb","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/final.pt","sha256":"e8535f87f9ff5e55214e7e9c0e260d24dd8d5a89e4fbab5b20726fbf7b94a77f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/final.safetensors","sha256":"3f879455c92c340bc3b7c8406e73ef8a756c0312582b9a3b63a0b0dd8f4a3d68","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"567c2897c7fed970def8a90cf6f6af8bde4652dc3d6572d25bdab334930c9bce","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/environment.json","sha256":"85fd93def055064d93196bb30ae5dd5b551809230c62592134eab526bb0b98ed","bytes":5197}},"requestSha256":"11e19b50caf4792738e83efdaeb464c26c3e841f3cc02e9ceb301c667f99dfcc","metricsSha256":"5942711d8c585408ec19c8f2877766dffdcd3a180f18b864335f506690442c1d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2450845","recipe_identity":"ed25802775448a9182f4c27018093fc501d9a8eca23b962407143dfd5eda81de","loss":3.5845053023134006,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/result.json","sha256":"ff643c64b6c4c5f636fe9c69db52bfe339d139e6fc0e652ec253c95afc40b3b6","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/final.pt","sha256":"5593a6aebd4d9c9ec2b2d480e7c9ae7e8edf0778c6158dbee42c4f331de2957a","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/final.safetensors","sha256":"584664eba51b8436537499d4cca951b84d623f91cfce12140ed726518d89c33f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"2e2d0a3f2be3b0c0e2a344e0cba2d132c532fd7fc5bd4636de923dd753443487","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/environment.json","sha256":"3729cc56cb5c960be3bb54f69a24f656a821351b769d838967b6bebb57122ba1","bytes":5195}},"requestSha256":"2f0dc210a910e8d5f218dbeaa536f6c7ac37ce90921d7b28d4c29eeb0aefbb90","metricsSha256":"e5a84b5c320efce3a15f7262e0d842998fb37eac57244fdfe851ed5df43b1210","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2450846","recipe_identity":"9d0cfb812e923b3af62f9d34ba3ba3ed96c81c26343cddd42b3869bc5f889cca","loss":3.5951866755998427,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/result.json","sha256":"6d929cad1d692f713e8e9c7a44193af5826592dba13ae25e66cf605b4af5b057","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/final.pt","sha256":"9cbb64aa070225fcbf79ff4f3c72bb841fd1032f7987d0088ac8c6433f267c37","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/final.safetensors","sha256":"e027658851e5c9cf1cb8f919f12d643df36d33bd2d7cb4a8da6787c93eb6b9f3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"e0108ec8bc598d7ddca6a5e7646eb60f6933dff66014496b1ea552cec090977b","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/environment.json","sha256":"21fdbbd7233729b9bd9cff1bef5d08ff37488279e41157a85591b65c361616c1","bytes":5196}},"requestSha256":"8c2520ca864a01598ccb26a5449d8e4a4069a04df6e732cfc80e37289fbccd43","metricsSha256":"93238967e39d65b038c18c65ce329bad46dfe617faaf74079b63c864ca545d3c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2450847","recipe_identity":"b22e5f7bdba2f6a49db4e0e688c9ecf1746f36a7fe25a7ea6fe596776e902e58","loss":3.5710420365123547,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"c494856f78cfce75ee941e690571260f0adedb902a182647382a8468086c8a14","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"e86ca71769cf30c8899b7ecfa0eb5bdec20a7116327a363b9a863b31854e91fd","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"e729d16bef51fa5819b0881d4054b76d07014a5bf602638c4485ccbc9e914238","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"02859cdc231d35f09d4aed4a1e50a9b268af006fbc21a22d62356e3afd84efdf","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/environment.json","sha256":"af87ea095be6a8a93e2634486b5810a700e83ecb5c5860f306c8d32d171e58bb","bytes":5168}},"requestSha256":"b3f12fe13e8f38b939bce6aa4acb45fd733892b20031115a920da7a19c4b4fd8","metricsSha256":"e66cadeee7bb8f1a34b679c210b0b39b47c923fd516b9eaefe0c7626406425db","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2450848","recipe_identity":"bc8307be1698222ea388d9ac28c301d18f0194ac1f1f13c14d20057bd392d328","loss":3.580096477316119,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"f365dfa763dc9d89eebab40f16684dbfa5db5a68f38791959bb29a8cfea024d1","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"3333d6e5f8d253037d116a23abc09a4bbbc11ea1e3af650687a80ca95bbd7aa0","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"ad24358bcf330ae2cc5a344ff497146c3a723fe515c6efa7712ff03be6e56f9b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"2647c0e4a3894a119274bf85c247661b859b247d826543405665671078d935d3","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/environment.json","sha256":"524aab173b7f4989e85d2f5aa15a5328d8307777839c9c53cc2c00768ce5af35","bytes":5197}},"requestSha256":"e3ed2e0b8e1997359a73e00db10405141ad1ef0c2da60b07cc98bab06c825114","metricsSha256":"d37bedc0ad477c5adc8d137c4a3c97c13fd19f3204ebfc612f76fe0103f60cfa","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2450849","recipe_identity":"0904b7b84d0382b171cbc6f3d7b9f6bf50775b2531a830606c24ba4611720f06","loss":3.5732184681253227,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"07037e2b70c011c97608a1930e3b7cd36bdc845711c1f570f0f875797d87228d","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"eee10f99b182d59e31c8f367a13424aa3a11c67b61252881c776abe6a2469433","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"761fe42a4e6c8416e606da27f338cf4687b471152e25900a0fc515a3c2adbb5c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"d78d7cd088ee789bc6df7c14778aa84e0487b17aa00494bd4383e33e44a2cbdf","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/environment.json","sha256":"1dba74654e6ce5a9bfe43df05eecc7447eacafb0a1cdcb6e0783747a82db6b5d","bytes":5168}},"requestSha256":"ed5a191ce2a7bf2eec346546cb05bf199ad9d841c5ee76837153ef81aef11069","metricsSha256":"6b5e050b513bd7e039c9453587b9bf2b98dc18aee05f24d89f167d229804ae8a","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2450850","recipe_identity":"54d6dc8e326ba85d05f0aaae6980ce6c64c251fe6a417dce217ef8375564edd9","loss":3.584592103854237,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"cfebcc21ecafff8736e2a081ff89e8c757f65644810a1dc042ad5ee7c0c1fe2c","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"f533de3fb6d5487bc796b046be8300e6d1de866e0ca793da8424baa293944089","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"34337c06267b1ab71fc42da95d61ee75c97df499336911ca1d8356f276d29511","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"41872da3f33c71ccf674da6c13c15f71a51b6190d53ebeb68576ab5c41d9fcd9","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/environment.json","sha256":"092f8572179969c1704f54369238f7e0ea26dbc3761aac2c18461e00463d6a01","bytes":5167}},"requestSha256":"66a8a140fc6d9abcd682d622d25fdc8b84872301085f9426875d2f1bb34a7c5c","metricsSha256":"8c490bd345e6b80bb90ebd0b1e2062e860a78ccfbcc4a9a323429e3466dc802a","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2450851","recipe_identity":"535472da4b118a09df64401538c9c3bd0d6d6feca8c65148300a2a95ad61f525","loss":3.5795696503595553,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"a909178b14a39916238b2132d375de316ba5e98ecd19487e5ea34b2c749b7680","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"95cfa67432e722363028ba38c8cd28c36ad8069fb49496dd9bc72a72bf1db265","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"5fb98e4fe9b069a808d503ca89aa37b748b21ea0049449222ae18ec69a035722","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"4ff6a74a03cfa574da179023fb68767c2d83adb899c401feb4a1cf417746949c","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/environment.json","sha256":"1f94cf3d3cfd6cfe17299a6e52316b5065f332b501ad81bc71f63b8d7d5a056a","bytes":5196}},"requestSha256":"c4c5a024abb28bb17bb409d28fd6b07d31ce58016e0e122a9d34d2062e100310","metricsSha256":"0e3ef9e0b223f9d01d2ee40a15b40014f02a0bae5ebeba105f1f9192f3d7dbf2","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2450852","recipe_identity":"5aa9ba7ee236df30fa8bec4ddd8c4262025c68c21fe5e626ebad2be6aad5fef7","loss":3.5820670418408516,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"ccc83cc2114aeb2f5a1d16cdcd3bc4507b746df04e13cf7f174ebc30f8f5ce32","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"4499f5ab266d8f515361ba79eac808cb8a834d3f066417530ad97fb0e9b82509","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"0bb2a0441769358dd60e19df825a9aeccc52f8bdc395b4c45c717a44f525678b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"01ae22a7bad332ef6b527fd1582f3b9c221fbe168e48bf7800db4108133063df","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/environment.json","sha256":"00bef2376ebc4fc4d2c2fbdd53531f98a45b1ba45f492b9a5fea5bd1f6604465","bytes":5109}},"requestSha256":"6813fd7fcc6be45d32e14d925c942c28e3784e89a9b5df051daf517d757b85cf","metricsSha256":"1280720167a5ace83ea32952fc8c314c37ec66c0c48db6b7f30cb087ce1a79cc","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2450853","recipe_identity":"800d011d12404717311dbeb5c3a4558cbf0faac0f40e68a3f656ddb1856bc9e5","loss":3.590969326333108,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"d6b331d435ea451b8dbfa750805853b6703e1fac67f6cf595d231b2cab1806dd","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"b09a5302093aaf79d218b2f3a2ae3bbaf1baf0a92bf826ba7d38ce42b2b7a597","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"66fc33f1c403f1a3cd021f37c83fe1e226d7f674263ebeff10c1ac1b6a494591","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"facdc80d956584f8442712a43e7ec61623544b80402bb343f02b3b1a195a3543","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment.json","sha256":"11b83f387763c606d112634284e58c2aa8f9e0b63e3463fa2345c86837af9082","bytes":5194}},"requestSha256":"30b9aa200a119d52de192299104276b63a5d0bbb74d5b221d03186be6fdefa2d","metricsSha256":"2a62bdfa1178f332f409e25f044d4f2a4734cb7cb8582c4f5799fc1438e90510","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2450854","recipe_identity":"ec856bd6402e2c28c29059d28d70c3c70786159194908429a57ea1dc199a4f39","loss":3.581787281399286,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"ab4ca7fcf0bb9af0e54e43acb0fa50ad5fe26c61ded48c2661c6087bf33a1118","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"b9ee7047d57905aaf357809043f456caa24e52502abf66356fdf9519fbe741f6","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"1e7635fa5a53bff643d8ddd650f2c102465e44feb34b98fd04b2fa16566657a4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"db583822e406beb943de02932deba7ae10fcdc084e7d0c2dcdba6664d3a62847","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment.json","sha256":"3da4851cacf4af49ec8b84959401997375a35076258c16c761939a753311e099","bytes":5109}},"requestSha256":"804afab7c7f84897ce3dfd47a69cf5679717025f1ca2c01db13d19247804355a","metricsSha256":"3139a1ea42192f423ae3401c7da353d02ddb0c2a6e0df64be41bf9b856b9393d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2450855","recipe_identity":"3b48d7bcf540e9099ad1d9331636ad6d2f1ff870131ed75d9a45ad9ddc7ce6cb","loss":3.5916640382110243,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"9786a28d146882787cd65c18c414ec167f339a21aca5fba089490ea7808fc4db","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"d0be415b9f3072257d54f57d6640fe2fe7ee2cb76cde9232ae1920920d3d5a96","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"f2f664c476698cf89c17753112f46d5fd3b371abffa94ddf9b8ae8c703dfb702","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"8a4f2db1fec408b0efe3d9316becd22e35df37ceabfcc840ee30599a198af0c3","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment.json","sha256":"6799267f0cb3f38d7ab40994a6a6e2353332a5265dfb442900b8061c299890b7","bytes":5171}},"requestSha256":"7d534f86aaf4a7bdcfe46161a0fb221302b725c46c5936cb3e851d3f430a73f5","metricsSha256":"b2f656bbf0fb40b2c87f4670557ae4311d198ff42e7443600495aff6db53bd1a","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2450856","recipe_identity":"9b5b93607e0d54dcd5fa03be06d5e503477b4d69e007b387d9286519a324d77a","loss":3.612617613381496,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"9c72f1a2c065c75933aeaec49d3dbf5145d83b99e5d7b773f32e54379aa7b870","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"ea085deb0c83ea1dd008b7aec4226b4394f0851b9d1aaa8be14bc1629ae6d02b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"7b3eb5dd634f000a17fc867edc5b2b53266452d099453933caeb3262dd0db294","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"c1737515403c4ad9c82b58eb273f761e6703367927e2df716856e364c8e466e4","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment.json","sha256":"f8f70039c6b91293d82e298b686ab527cac11f8559d8cf6c8e7e3a09cc8c217a","bytes":5201}},"requestSha256":"798936a2f8437447c34284a8e3390ad622a8411ee165a284b89b5c36737aa285","metricsSha256":"05e82c8846375af2ef7741c437d2bc235980af48df4cc425d1542e56d277497c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2450857","recipe_identity":"fb58acd1e3e8a31c52ae65d7659745e904a95f38da1c39b562959add19f17e9b","loss":3.595483762233382,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"f6579974c9874dc35dea9094ddc737bec79090bcb6fa056f0cbc64a529ba4532","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"a70bad3f36df61ebef81cb2921366b7de4d368cf4d00f12a7c0a6736ad2ae041","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"c9580dda15d3f03f19c968a522b5d344787b55f3cd9e03ff7e1119ad8bc3a1ee","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"b3abd46c3b78afa17fe7f9d2ae949e1ea2ce64e7373f33ff6e7fadb4adf236ec","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment.json","sha256":"a1c5c8e998c6e00fd618b6aaba4a84f21464c2bf9c8aadf7070ec9cf384b75aa","bytes":5201}},"requestSha256":"b105da690212c5e35b8d61fe8affa089f8c4c46f4b29db58909a647f2f9c8486","metricsSha256":"168fa0c70005db45d8338e47f8a31adc65978c500af289baa4a12f7415702a63","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2450858","recipe_identity":"7e33810613fa5bcbf16e5236d958221dc951db67b3b27e536a9ba77ef1c41851","loss":3.599385396911696,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"1bfbf1bcc17faff16efed24763397040d7256faecaf4d97bb93efdf2f70dadd2","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"0c772aa5b85bc511bfa3a20399c64d5bb0a29f4e07f998288865c4440971fd6d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"485c3d0323409108c7e10ae1aa528b69ccc58ac9c6f82ad2747b9c210183ba9e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"0d020808fb733c39655a2fd556644ed9620f7e83213794be7f6731100c5cbd94","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment.json","sha256":"01912c69282eba7d26bc96cb519b553a2074a03b8c80dccdae84727631179761","bytes":5115}},"requestSha256":"e4fb43adc6beea7d0b98fb7174f1e9c87eff76a10eabe9b581f2203836f2e71b","metricsSha256":"bb3209144e7e065fe426f3eff66fef4c34c7ad4b2a9a1d10623bb50c500e493d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2450859","recipe_identity":"63edcc2ae056f61987339bcdb23b26553fd49d83d10ffb9b45321245defb7335","loss":3.571040624847379,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/result.json","sha256":"22f02269a18eeb8a889f92e31f5f6dd579198caa80db88e6186dfee3219af9be","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/final.pt","sha256":"7d6f9f1f7584a11f3e725fa708665a5c7d710162c6dcfb98caa6bd0dbf40d2d8","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/final.safetensors","sha256":"a33c094c9a668c8da61dc63501ba1a6c19bb819e06e670a13482ef70cb671f08","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"33520dc0742fb8ef66a5f7091c21eb70826988188306c675791cbcb766f31f6b","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/environment.json","sha256":"c7048ae65bbd8fc7c0ee6f8230061bb160bf28f2e80559848e4646cd6c34bfbf","bytes":5171}},"requestSha256":"b26ea2d858a4d389afdb81a8b61c5f1759cbf6d8c894905a98ae2b5d6c63c111","metricsSha256":"bd6fe8f714da03007b16c8c43b7dd6ea0fd61e3284610fe9250db8278973433e","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2450860","recipe_identity":"f1b144ba701426367f06e44590133b145874c8f4bb952a46ae0218b94454b139","loss":3.5802491183136085,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/result.json","sha256":"095c64fbb8455bdcfd02d7a1705166bcbe0b5a3a602cfa6a761d78527e078d2c","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/final.pt","sha256":"43811f96dbb47e01f7a9e72d335ab661eff0731812f177c2b79e53c969cf6428","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/final.safetensors","sha256":"3db1d3a3de11d30fb81e4b55a26259549fc59f87dfd98300cd13f7e8ee0309f8","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"b4ef9a67ad5e8bd979eaff23559090286d1f13597a1edb61089e7ba9449e8712","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/environment.json","sha256":"8a84bbcbc1c882bf68c21133bb7d5104a8f3e8593986bc904fcc7d56257dc3af","bytes":5200}},"requestSha256":"8ff1539c5dc6997b8041f5d0bbaeeae01e51c5580203fa9a2aba3777b629ee0d","metricsSha256":"6f5c032ed9321fa2bede6d5d682cd313f4b34293a4dc264aa47576497acd6c3a","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2450861","recipe_identity":"8a60ea464f96f48217d7ca97e639d3c14c586d4af069b844be63a236763bd42e","loss":3.569550507584654,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/result.json","sha256":"0682ec4570a46b704d47a8c955425aba57d13664ae8ddee670ac9284bc647c6e","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/final.pt","sha256":"720927df1b19ad4aeb663fe21087be5f1cc6a18f11b5743b9b4bbfaaf1cf8b2d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/final.safetensors","sha256":"79056653ff6db8a199db8bcc3ec3ca2621c4b877defc2a8cb695504e086194ef","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"3580c5c21216c888103c2eabbcb4353796d96a43eebe13e1afd0a363f55fd526","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/environment.json","sha256":"f1f58c929f6be9a7fd8bf02225314dabe1b6c531e531d94add5f0acc9d92585f","bytes":5200}},"requestSha256":"64c9fa04eab42f39d63ee7dcccfdde051c2e3961c12668162655b680f1812686","metricsSha256":"4ad4e54236bf8cb51d5f8cf25be7e0b33202a5190a6d1b18af76c051304d4b3c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2450862","recipe_identity":"df1cda8f98b14f62dbb428fd1201a7fd78ca03e55698014a1340ee87a98b417a","loss":3.568652845270944,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"dbf573758d0cedf348ef826db9aaa2a8478e8feca857f1e1b5f314e38be215da","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"7f27c1f47ef3b90f400a05c1c77809768c878acd14123cf3364bb0e615e3f253","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"4798ff7cac6751ca237cde05d7e07b95bb5c1c63b492100a9caf924aa19a8d89","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"0e37ab6e06e4b4aa06976e4aa9434a3b2eec73c6bf7c66f554009c2585ef729a","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/environment.json","sha256":"51949f04c2406c3cedbe3881aa613e2960114a19833f3fe3c9126f6c1bcb24e5","bytes":5171}},"requestSha256":"b4eb720b9049d54988e25487111fff34fbf1963daa278233536761e0e0faf643","metricsSha256":"5d936dd4f19a5e833ff24905b88d9fa3ed30c45c3f2c2f8df9d5abd5bb47e46e","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2450863","recipe_identity":"c55f4a894e3ed8389e4f252e078bb5e0c8219d38f17e95c01fb689fe77d5903c","loss":3.5643524773799378,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"bf282ca848a5cce9af092635075b6b1ce56c825ad5f01c8ddceaad9304abc12e","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"c8c848daf073bfeb9aa9d5444b48cfcbc79252f41d5bb49f02c9389d0a620256","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"f54c66202f8a9b4166d08c609f6c670786809553ba6ddeafd8585ab590e7db8c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"d21cb28c35104b971778abd310355f9e64c9d97ec9684a39195e43a1c940cb84","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/environment.json","sha256":"55889acad933167f6a6f9a5ad32c02fc211940d32da2792aa73b0e6af2b11c64","bytes":5173}},"requestSha256":"2ad63cc4a23f43d3d2d3ad5ed738b9e72c7257ebd14bfac68cc908b9a85d8381","metricsSha256":"115610b88513ec52ed22981f9b6c42e451794a6cdf482b23160b8ff90fbdad60","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2450864","recipe_identity":"51658740bb7cb3d9499bf5686151fb5beb1cfe3828871f7628594c1bbbad1231","loss":3.573471836063833,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"89e3675e917e05a9d50c21daf9026d7a741ad18c014c99b10297755ced10fc79","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"3cbeb7e6af66c4bac490c4d3855e598995c8f244850890a3630813931f5c1f8b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"70830bcf62a5c83eb27f15354bec06a90932d20834d110bf43e173c7c3c044cc","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"171b8ac2fa2af44e94e4c5e331f41856838a6a1d79ac12cb79af46310363bfd4","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/environment.json","sha256":"00cee735abc6246fed9995052dc87b521230d22ee3c4f9c812fc5ffb24e4951e","bytes":5115}},"requestSha256":"8cb3fe045a8b00d811bc4a94596886b60f7ca8ae4ec7568f65d572a40993be37","metricsSha256":"69b3cc6f2c60151fbd9aa2ef1532f143bfa97110ac0a84f82115505b4a9a86ea","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s42","horizon":20,"attempt":3,"job_id":"2449510","recipe_identity":"fbfac5d6f7fde07f4d3a17d75cec051ff6bff880a43c9bad8b2b816e955171b4","loss":3.5599536061188073,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/result.json","sha256":"c47a118e67457a9d136f06af30f9178363bef80e18911158d9aec13ad48f813f","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/final.pt","sha256":"757153ccff05afc8270ab0a217cb062e9ce5541044320073d81da0b44edd2923","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/final.safetensors","sha256":"b2ef881b24d5441e176d8200e874b99565425aa377fd3407c38ea61a24a210b8","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/epoch-035.pt","sha256":"7ef4456c4f99b563768690d3058be4c5fc5d2ff4e8f7d44e1a7a32523458557f","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/environment.json","sha256":"3f0a1e8444fc4217a09f69ebeda02254b8b5e2c1f4d4fdfeec03a3956041eab8","bytes":5195}},"requestSha256":"9e9b9f20ec08879b2090171fa471863d126f981b96350695554affbe8b0e5b85","metricsSha256":"9294cc01b6e6a19774e9d0a451405aafc6eef6b7a4ea961b172ae7d556d4c4f2","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2450865","recipe_identity":"244a0a4f330887756c78df3e67dd685586c23e8576d01dfcf90e908a03d10f1c","loss":3.5779362573653644,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"f1c62422fc1a4c2d542f30f42d79ee9b30e6ba6c5278f044444a7dd67707b86f","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"1709ffc7588eecfd34cf5c8d571d773986b15d6cafc297551c1ed2fa39b8a363","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"2189fb7cdf01a0eb3cb79bfd752ae4ea590ad640959b3141286f07d4843d2a44","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"c2a9cccc0002d653609bc66416f2051c85f03161603d3c5e6ea49a72e6b5cb60","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/environment.json","sha256":"216d1f4014fcdabf2f9066e7cff6ea0d991388000c5c26b5243cbc4f97cf5744","bytes":5173}},"requestSha256":"42ccad6f001389276f8ec95d3be3f4acb07321410c388362a4ee67059c8d3a33","metricsSha256":"76bf12ebdcfd7af8af206a713aff8fd48ec1dec16c2fb83ca787c10a2ada1d89","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2450866","recipe_identity":"460b5b352c7281894b7d5e954b964259a6982885dc908b1291237dd3f3a33ccd","loss":3.576087391373635,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"73b9bdc4518633b2b325c64f17338832802d2c4933930d2053df4e096322f50f","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"e43187fd81a411bf184ea0190bfb53102690bc134ebee9689de70c86d75289d3","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"965fd1c8514d7445d3b58d1f98f0213ab30b4386762fd64b207059495af0b459","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"2ad79036697402fe621c1d8e0ec45d1a2d66ed27af15be4c088f590934251480","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/environment.json","sha256":"971ac03aa4b8f52515648fa9bd3eba43177ff5a2c47364a21adb6757ce62d89b","bytes":5173}},"requestSha256":"e80189b3c302133aeecc8f30d2da38863463be446d1e57dbf96ed3e02e810333","metricsSha256":"c27a7f317f01f9926c6735fd3da7c007ae79222efb2676a61c1a96f4b34293c0","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2450867","recipe_identity":"dbec4bad17a92472c0d1f38ce46ee688df78d35aeb09c20233b9b3365b2163d8","loss":3.584645328549556,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"9390defabeb3f77bbf33a7449ca77fbec12daaaf754183dc1cf40fcf465a6a10","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"e07e13be51da648d3c37a282831069821fd6f8b3a90bddbe96229da5f90fdc7e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"f93f5e538c6bf25971fc54542c10f0b14e4d3b06633269e5513933ff21593224","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"2aac8d83b1c08a3863bfcae4200ee1c569b06006c75eabb33ca98fec25b84c7f","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment.json","sha256":"5221fce264c494946ba82008f44b2f275419d281c2ce93519b455790fba6b726","bytes":5200}},"requestSha256":"11e45007dca085f8670abd0c69d2d3fc7dc689459e74a13049dddfdce3001653","metricsSha256":"95f24fa152c83e80c470903064c7efb9fb2b99840b617da86e1ad42f39d2982b","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2450868","recipe_identity":"e9afde9cbafe9533ae8a758d741bac74fc0a95201f617946ec567ebd5b076a11","loss":3.5786623300670466,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"daaeee183ba2b031a739bb0b744d9e2dc31a98847f857845ccf3cbe41b188ae3","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"a92aa49f0cd02383e326f7f0446f14ede24075fabf276bb304f1a8fc4af81c0f","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"29f10b99757ece49a418bdd9714845e72b3c9e7bde32725ec4f48fcffbcd104e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"d659037d3dd282abb4a0e8879ede0cfa61ef738ad42d550f51103cf8287996f8","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment.json","sha256":"8426da76f3c4fd4ba41c07ccfe172939ea69a2123951767709b0e1a3d24fde74","bytes":5173}},"requestSha256":"ac027941d1ff8450f39bd88cec44db4b715d21b884960a14199867a7f855e2e8","metricsSha256":"b3236852f9292bf1b84295674dda04e50685dd8268134112dca81e8646c1a147","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2450869","recipe_identity":"f64cefe8c9e4d238872951fa098619e49fd72c6e37f2a12ebf119beffd5031af","loss":3.5863996854865703,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"593360760bd88571e7557f8ceff9de7e97614f2c83b79217ccd0e2b535c65639","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"c91c8de5acdce5fcc4caf1c18b0a12b5bb15a1a0763b7d4fb76429291f4eeda0","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"680317b65d28d283bfac7ec499cb4b722938c79fcf649a0b6c44af79bc733d77","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"0fa944632d7ee2a955d762d5139ff673aac2cd1877358bc684027297aaaa44a3","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment.json","sha256":"9e16446b2004cc023dc6083650b15aaa7aff3c20660e0c2b8446c74de2748988","bytes":5202}},"requestSha256":"0981d470e20266201aadaa667a6264b899f0cdcf88b500efdd5c8602658ae21d","metricsSha256":"2a3a0785b10689c776c75a4928067f2c5f4642f551310a209c6f49828aa2e1cc","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2450870","recipe_identity":"bfe6f702a4e125e905088e5864c21d042176fde600002e5632562c7904149cae","loss":3.594647453114633,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"2d939ab4dc4efc4217cbf69a24910c0ec5331b1eb3cb5575ce872373103f9a3a","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"3fab7e9d91111ef31ac5635731b9bbd6ed801bc6b62eb282d8d9d0d9c9f50f72","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"b24c210dd71c1d8a7b0c94eae9664385f1d24bcc311234ba6aa4dab748d84973","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"054982a61b8d9ad9154217d90cc54bf83db2513df72430a3f7bd68bd56554b70","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment.json","sha256":"0f346eecf42995022ce18c386d2eefe3c61fbb8fce743148e0ddd0308eba5fda","bytes":5202}},"requestSha256":"91bafe5add7ba9a2b4937f90e3c42dcec9a52b2c2fa23d45d5e0d33b1a930131","metricsSha256":"02555c79c3aba8a941e2bc5df7949b09ca6c42cf5c788e2c1d9ff71205c6e835","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2450871","recipe_identity":"59fc08d263067e76cfda5d21298061c03a966803c1a315d06147b737d2a6ccc7","loss":3.5819512454868043,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"4712ade6866d663270bdbce7bb7aa54f8dfb4b315515cdce0f6bc5453cfbe67f","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"900ef92b0a0fb082c3eb5bfb21cf55f189498303845ae52372aa732a648d4f38","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"9acf2f20f1401b669eeb2dc9f9f69160bd15006347845772e88c57cfba87c0b2","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"e754c46bc32d0b8f5f9a56bd680a627c8a0f6b0f89a789c4012f8017ba6b4b06","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment.json","sha256":"6946653b25c0abc60e6ee10f50d445c22e358fbdb96b8f94e93b2fc287da853e","bytes":5113}},"requestSha256":"e21d81b7aa2565af1020880e4edca8179e77f648e7b29a3054b2336e0a6486ba","metricsSha256":"eea08a1aa14318f4bd61e9a4a605edc6bf2d508c0e2c00af67033364e9b4495b","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2450872","recipe_identity":"30d49c03e5e0ecc9c431f49c252bbf98896d26e41f136d22d61b391045880886","loss":3.5940430049418595,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"d5532480a6b1956dfa4e00edac0c1e9b341df39c0f9a90aaa8e86ceb2ebf3627","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"e94e1c509068a04b1d7252c6eec5ffcb78c35b0556a2caa7a786da78a13db938","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"cc3295a9088e8930427a0dfd412e82393ac901a1e806b041aeb3d2de2acedd7d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"85612c22f2a172adb68eeda4581dbc5f6e4e91f17360770bb8b9d82d5172e7e8","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment.json","sha256":"4c1405d2510c637de133a8dfb73a3357f2682a5864ea86f9cc6eb39d062391b7","bytes":5200}},"requestSha256":"2a7928b6364873e7a992ad869c1dd96169538509084c68715a10c9fc7b04f5f9","metricsSha256":"b8a7e00030bf0701f591dcccadf8dcdc43d800f505776b64d45f6d4343b856c1","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2450873","recipe_identity":"8e83a979a769f054a02a9de0d7970b8f06c9f1d22b6c384c69a83b29c5f60f03","loss":4.278052927002443,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/result.json","sha256":"cf37fd6490f009cd856e7fcc805ad63ad8dfdabc9651c2f4e8958eaef3e5fe96","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/final.pt","sha256":"ebb1d4946ec037c9beed62c1f59eaf89d7e2fd7e1b0a5ba748eac91734cff10b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/final.safetensors","sha256":"1a6606d82e59db17d451531c5becf911ebda869a06e2022da8f5e58cefe26ec5","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"d0d5bb7793ab4ff74e86f5dbaa83b7b92717bb601b1032afa5950597433e9fcd","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/environment.json","sha256":"3ae77527019a53b1d696f3b7606587717eab569793be17d6037ff24880450b6e","bytes":5173}},"requestSha256":"af76be984f45300b924e27fa752860846df7f7b2e68dde3649e7959164d4c4f3","metricsSha256":"605be89e05474d26e5f927dfdbb6a0ef7b90ebd8e5d17dc8950b4c520ce67cdc","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2450874","recipe_identity":"f8ad393caed352eef89c3fba33e927c768bcf321412fb82c50dad78fefeabb6d","loss":3.6021826013279674,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/result.json","sha256":"292aa363384183a7e2d55959fde5ba52075bee8bc1af899a65ac21e417a6681d","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/final.pt","sha256":"08583b9455c050ed34c26f2f1b0a07bca6fda321db425dfd6158cf4087e12cae","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/final.safetensors","sha256":"fea93bacf0e2a1c067944cb4efc7df2179a4fc5e1d985bcb519e5893f0779e0d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"3498de97ff16020472875f67341e455575a5a55cb68440d3beefa33bd51f10c8","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/environment.json","sha256":"72700ab9ddc8608e07f9c3f2c6284618cd05ecfc229d166ad329ebf935a7ecad","bytes":5202}},"requestSha256":"41c568d877a195576032a224c9a4e7e6f3f3b067ee42a38204588be33f69d00a","metricsSha256":"cf9bb93f1f2676786069d12f1b62f65cbddcf99a79d41332e23248f859530648","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2450875","recipe_identity":"dd5d5db2cb610c1ad787eb008c10841dd14de51d2520ced26cacef7e15e2393c","loss":3.5910854813113016,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/result.json","sha256":"7526b6140e50726b2de522253b9ee152db6d67ebfae9cee171c993bf154ba97f","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/final.pt","sha256":"60894c6899067e77f7f9d80038194af65d6a23b9af2662fb1fa154678ec42afa","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/final.safetensors","sha256":"c38599f9298def7be0ecdf78c7953d745b5d066d4f0da664843cc309098db233","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"3e09fbc80afdb2249e4ad69ef3f0603a72213b7462d8c2f07891f682446d24f5","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/environment.json","sha256":"5e25248e04b80d4394bbc540d866bd3b3392912c0d5c3eb8618cd2c0b714e096","bytes":5172}},"requestSha256":"114fe03e72488eb16fa2d23bc5930568b48bfef732ed9f2d4c09dc23aac74516","metricsSha256":"0e3b13e04805304254355fd7ca3b92c33e18532d2196da1110b959a63e346f54","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2450876","recipe_identity":"55fcc489cc29e17130ad11cba95393adc475b8ad6524dd9effb94485bde52e76","loss":3.576374475218168,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"494bed1759bfcf8ecf8c4f29082340f79ba4759b0ed726ca40a13dcc68663276","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"da5727805696a47a491f59a37b73886c64b4baf5f5271ba122fe8c916f2d42f5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"f1fe5baee014c44640538058a72c387f87da528008fc71c6df25be59a84a76cd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"90da8ffe535b425abc604fb3611bc540f236a4c19bb44f34ffe6c83ae7b67112","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/environment.json","sha256":"38e86a1c1295ab061671d7b1b812d8116e5aeeca3e9ad8fe2e4ab195094b17d3","bytes":5201}},"requestSha256":"067c21011242e3259e9786bcb114e9febd24cd9a9fbd8939c723ddca2ebbdd5b","metricsSha256":"955c4cd4878ec15d8509a4131a470129089e82e63a12e766ad7d18bae44eafb8","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2450877","recipe_identity":"1d55fff59934098db9320882537ea06b51eb205545e79c44d4cf3328a94852c1","loss":3.5731031750731845,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"97ba61b3649cea69329e231c6126e87af4b80427cf7f18c0f912f6b5f4c227dc","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"7a6ab90cd5cd5769153da66369a87c68f877616d2b7c35a39c2bc2b16f7ad378","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"67fc7c8c2a3db6aea4197918baba9a84599f602b62afc4ddecce0ee8a73a0a16","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"41561dd095b00df720d56e3711346e7f1d33dbc4ca0e0b847f8f116d12e81e48","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/environment.json","sha256":"db22dfb49dedc77db980ac6e0210d81f105e77c4d0ec99886163d116e74a4657","bytes":5114}},"requestSha256":"0d804b20c88fc1d808424f4fa327ed9e154df865bb9efff6f8e7231ae029315c","metricsSha256":"583caa373d8ff4dfbe71affa2bee45bb18f5c84dc7434ed2bef7c1360033f57b","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2450878","recipe_identity":"0599194c104e352d6d125f58770830c04fc142c0d7f55bf36d2d91e725551229","loss":3.644299475739938,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"21c44759b17d6ea00fccb7d737e6cc012cdd43a569a5a8d58b896d491e76a3a4","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"4684cdbf91a15de2124122491db3273c294b5e816f492c6177c138f2587741a8","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"e2760111ba0470054135fc0abc10e9230624fe08d621b20ef398c345ff2bf49e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"dd868c1f4edb506ebd27e52dd3dbe1060f380450d8ff7e9aa924ad6b93f30894","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/environment.json","sha256":"85420f22083f4fba8e17b999b6bde6b490c14bfc5e263a41102ba0b1a623e3fc","bytes":5169}},"requestSha256":"cef8ab7b98ed99ca774765b24b79d835427c603fadc1e2eb0060bbfb6596e897","metricsSha256":"c57afbefb9a7fdd13eb2f1bc153a7a5931cd5416548ac7fce3af4126a8b6ee9b","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2450879","recipe_identity":"9b28a46ecb73b427269826507c2f3e3f167db0f5ce2b5bfa5d58cd857f4732e0","loss":3.5796302724512543,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"7c4707d62e206668b64dd97eefe2bbc320bcfe814bdc607bc5d59a6ea922d595","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"4812bd358af336d8fe4aa5f31cb5b8aeade1bf8a3d79d899c86d0de3384115c8","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"608fda7cf203961e6d5fc309b9458c9bac0733b64e8d5e1e698defec332de9aa","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"28969ba28acc8d42f856203eb6028bed4552cf1f6954d561870af43547a796a5","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/environment.json","sha256":"567d8742fc2a601a696070af0f26825fdf19845f19248888c3d21ce4a8bfafa9","bytes":5199}},"requestSha256":"ee9fc1b2ed37da3804a5fca65beef88faf26dc263e50595dc1cdd3b72676335c","metricsSha256":"89262683ccc00e1f91e5e2a9355b17aa1a40270e4061a1d0b3000df50943355d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2450880","recipe_identity":"1a0820bc550260835a8d2ff245ed6f7262ccc130f6a3293a712783e94501dffd","loss":3.5838644489429505,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"71f2722504322b924550028c0faa8c73737f1ce1ca3e75158b1ed690ff53c639","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"802e465b0f5f6e8ee0782dd9360682acb9a2d9bcef8790d83685b7cac9794571","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"3c586f7a0b8f972c46123561d2d9a0b5173488f4610d0f1ed5dec56bd64bad84","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"26e4a43dc2d6c902852650c8a81f1b3de5e14065c68f9bbbcaa156127fe7e0f9","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/environment.json","sha256":"7acd38b286c5a02f83cff13bdbbcaf4b88dd11f062ebd45bdb2b37cf20143bc9","bytes":5114}},"requestSha256":"de318bba93ea9f6fe4ce6870c4d8d82cf73a9606e2b000f4bcefd4ec1ac37d06","metricsSha256":"7275d882e7186558d93e5607017eeb2e48b803003b60cba7e1bf98db0a77950c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2450881","recipe_identity":"9ed724fd6d66dc1d270cd7502700ffa9b3c6ce23592cbbd592f1cd0f492995fe","loss":3.575918023174999,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"b7eb1ca932bbe12b216ad1101bc8f919724d8bab39d04dc5cdfcd189f564a132","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"19d2afa14a0b523086a85c08b00b12232bbfe64a238680394c0c9d1ba334890b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"b8300701eb26f8c410c4e3024c3ce682bd1e7bbe82b94e55e3bad90ccaa5d06d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"59c13acb96c4f6e68a8ad44dd904531a79c7d05c5e4dd585244b6b9d848ddedc","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/environment.json","sha256":"f90dbb55eafcf4a490e1362dbf736fa2c0617150fc454e91589ddb8cdf7c56c7","bytes":5171}},"requestSha256":"dc4c41e6b965d577757790e5c64784d5672faa2cf2741903fd78a5de5e074e40","metricsSha256":"69302081f6ea5fc4971d8a23b35ffa25bc0c64a33e582eb8699bcba61933491f","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2450882","recipe_identity":"da53fca47fea2be4fa03b06a5ab1d2a3c1be3fc854325cf0ef670499bc7dc593","loss":3.574305882349551,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"68a22393546a4f1ea9fbb2095a0dcab1a2560d811d533c56e785a63673b6e303","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"1ac05139c9255d03e0eeb68c14a4159d518e081f64b62cfecf8cf0ffbb8355b6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"274f86dbe044bcf9f56f8689feccff9b8e72db05bbe8db947d2ccf5e42025df7","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"e978cfcda23620d19ce3f8330bc41252bf3b3ccb2b52001324078d1ffc9d319e","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment.json","sha256":"db4a6cedbb14dd77daca1c7f79afbe1490b28262e703990678f11866d679d0f4","bytes":5113}},"requestSha256":"9761e2215fbed33715b46463208b0f0379e109a576f304a0c558ff007bb9129f","metricsSha256":"bf47f42ad9c3ba7fc4edd9cef183319f0b479c75b8d4296a26ed876c70fe61ce","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2450883","recipe_identity":"840287bd4ea4758f204c763cae7c0075e49ca3944c801fa18c6f60d883d4104b","loss":3.6011489677024087,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"a3e8cee64b9f4021a0fe84fa00f74fec48d3e5be2ad67a6988b5a236d7031e1a","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"71b971f448e0d10de155a85689b6bdb09b4d719493e9d86f228143088e9a6e4e","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"b0349a698590e76ef78852bd95afb6584b217909ca63118408e66e5761a274dc","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"5d1e0e20001eb3b637269f7ce46d517baa448d76f91bb39a65e9ffa16a0c8cef","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment.json","sha256":"cc6e56d5c1f1c6f26e507b6902e523980ba8fe766d41ffc9a72d8f032b1e4f0c","bytes":5171}},"requestSha256":"47eb5b11981f697145a79b79cc15019acf84b663ed59840cbe8a650c4c4d447b","metricsSha256":"4d37c28665c4618c516853691324657ba3ca23c25fe9c0336791cf7d0fe76524","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2450884","recipe_identity":"5aaaf035810427de61588c10be5a435509199a408ba6f031ca30b99aef4b335f","loss":3.6022701932327457,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"6e8efb01695948b2b5b8a4abbadbdd0bddf68a23cf6cfdbc82bd6c927012a8f4","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"ead21f292c92e0f70fddc6627161d31be4f999605b66c0b168f84054964e1bf5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"f392aebe9fdf7fd7d01f1d4405a0febc9433d729c33441c17a47665657f4bacd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"037d5b34fba722819ae56f1fee9e773e0efeb7de869f9fb32e91aeac83369bc1","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment.json","sha256":"90ae1693bcabed426aa21efce4994f296d04f81c8516e0d972ab2a40ed9d0e03","bytes":5200}},"requestSha256":"67886c78f9957004ec37b65b67c93c660c56e883a4c66e8a5e05eb79ad2a8c11","metricsSha256":"3e962e4cd9c361fbc7c527a51437c645433eb19672007de2cae1e47f8ffc5219","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2450885","recipe_identity":"5ef8ad14a4a8cb6eb89a7f625c7e3b71a5a0bb4e5c40a32ca0881234d8fd3709","loss":3.6048555121924224,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"1cff457a8b5fa9aaf860ab764d39c14a93aee7a234deb9b1c2101b3ba123c870","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"3bd86f6ba476367fb32dce2ab8a9311ac368fab7d10a904723e19fac59e0d5b9","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"b07e392b04f54ab8c9d00ce7aba8100911f1ec8037fbf1b74a3285726e9e32da","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"b3440ae8d3c966c3727d49c4889eb570eebb21cf69cc13168f0d70dd2cda006d","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment.json","sha256":"4892d419dd79a29ed204a082f56306140dbc415f514039704ecaca536a21d1b2","bytes":5173}},"requestSha256":"06c880986366288c2f5872d7d86d95a09a2201475320f197d409c559c310a29b","metricsSha256":"8f6cd5a18d9adc0e28a25adc0aef0bd794ed349cfe02e22f5509be07ab499fea","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2450886","recipe_identity":"0f8cfc9dc229ba0cea536e36ba09d8c5cba7bbb5a6bb2b340c1b571ec1316454","loss":3.5988897075367965,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"4f6d623afadaca73ae1901ef09fdff46b478f928a28d0dfba7dc8a02246d4562","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"35768bc8da9566b978519ebd2dd38ae9e18a53c03b48032ecd738d273b61a724","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"6a4496eca28072ab76e3bfbb10f74f58eaf8d16de7775220567a880d9832f225","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"24a2999ef4f2009ccea13ae49b38810994a6008ba27b884018f73340ae0aadb0","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment.json","sha256":"d8a2df7fb43c50eba36b928e6584fb7d1fd398b1e3c15ec172dafce6bdc0accc","bytes":5171}},"requestSha256":"9317e8f3419e86c3920d2325dfcf067b5772a47ba9f2f5a6a2884ebda3c197de","metricsSha256":"5c7abd6653c21d2fbd70562d44c12201d393ac7aefdac37f8ae359949008298f","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2450887","recipe_identity":"f92dc85f08c6a7a690497a280d6a347088489e34cc7a454e352a22731b12005e","loss":3.5969445118891783,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"c52f0a19f0d6c01cdd907dcded85803164ed65c4b71f6660d4de4d9e7a7e2752","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"06c86ad5ab8b43fc92646d661572f2a94d1bbd7818297752098f47c70ea01697","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"fd1dff757bac2cb29f5a62d6106a8daeb846fda3005edfd7acc4c9ad7b0a8604","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"ab03b960348588593cc46b62b351f6cf43e82c686409bcc07f95aba2846921c6","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment.json","sha256":"b66cb9c7cdd0f90c54fc7f44d5ea6a0facca872b38aa08f81e803f8213f13802","bytes":5200}},"requestSha256":"746093f012a2c2c4b665658bda31b6a7b0e92c3a95649fdd211bd86e255aeeef","metricsSha256":"2a52e865b6d6c36877ebac7758ae110d3b8972210f0fb59b81c0e8a6351871fd","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2450888","recipe_identity":"b2d2b18b3ffcffe75d9619dfd196ef24b83fb3dbb46a00c4033e0d6b698b38f6","loss":3.5809341788065727,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/result.json","sha256":"6d9105c9913c4bffc3e8653abf52a88b476ef9a3aa00bef12ae1a0b17d225040","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/final.pt","sha256":"e4a464f2c32ee0fc7731de3ca7a3c046bddb751004e30febac5b8c4cd8f8ee9c","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/final.safetensors","sha256":"e7d510885abad11e0da706499f017778918b245ae85799f88dd6ec9f2502ed6a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"58579a9bf5f868efa083875b385e0fc1ef54686bad8dc2b311041e801da9a1ff","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/environment.json","sha256":"c720d4c4c118bbcb5be3c155929f7f5830ab7a00a48d3c87ddfda5e9191528b4","bytes":5173}},"requestSha256":"f2a69e859f6558315302467f0a827b48e6333cf53cb9332d68e02d02cc1c2eef","metricsSha256":"e0df4a0c933d60ca6104dc4cf7e0af084b79e35c75a878e875b467d059377c1d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2450889","recipe_identity":"78b041759ea45deedd561a1ba47783100649b4265a455c0e06a3d59a8dc47d18","loss":3.5790427700718372,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/result.json","sha256":"89aa70cdf76e6ec4b6bc6035cf15fd132303c341d129bce862c0636a98c43ba0","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/final.pt","sha256":"13b6a0445ac077d0b7ce13a6aaa435a91b41075b72ff3d9048ff169c81c95484","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/final.safetensors","sha256":"e67654cdeb3b69f77b2df0cee827d9e9bfe7b919d57d705450eff8a73cf0fa7b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"519ce05c89851b3458879cedadf0296560d91865092a3eb5b7867a686cc4916c","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/environment.json","sha256":"5f4ba5188255b5540ef19e69c2f7bf65a862d39cf562a3b16af763f79d67ebb9","bytes":5202}},"requestSha256":"88a49809564c7ca1424b1eb323e0a6d7e41b0f2441f8c15a37ade062ff755fde","metricsSha256":"1f2dda77b012026ff6a7088080d277f924ad9921dcefb8b9068beae74202ee0a","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2450890","recipe_identity":"abef7acb18acfdefef548e8ab6f9845c41a931401dda81624122727f07cbffe6","loss":3.583277486065434,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/result.json","sha256":"de431fe941fe1754bb14f3e2796928216428d71112246db748ba19bc79a996d4","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/final.pt","sha256":"96ae0f7d7e7b389ac31ecb664b31aee688779306164b111c3eb522dddaa8ae9b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/final.safetensors","sha256":"a040cab1245581ea91d0789861e9cc0060ad7899763b28a374bcc3c6e17f450c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"af0f476897c27fc10e4fb6b300b272359f9a56b4a2e63daffb0ad572dea67f0a","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/environment.json","sha256":"9de4ea0b18cb9e22b1f04de581fb1f021902e13c91f4292baa1937265e0ca1b9","bytes":5196}},"requestSha256":"25f912f9bd16ea7e5f836f336f8df55799af5bb95da25ba833539303fca97916","metricsSha256":"43609fb2de8cec03d566fd8adb1f522ee9f09c00520cebb2af6ae957f5553251","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2450891","recipe_identity":"23caf036a37e08187fd15cc37a6ae74b8654817763d6bf4f92c6e6194fdf85bc","loss":3.569095058628378,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"4e58acbcd14d3f18317e4f10cf7e97f23a96975386f2109fbab57b505c264698","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"90d695b841f02031987527a1f91132f81d8be6533ab3b8923c1330b62d762501","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"96ba3e14844c757648475f710bcc8b4950292e44c7d4e93a05a491dcc3c42f0e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"306334c10fc4bf37523e530bd2cb3bbd7608a1a89de443fb5054f7cb7fb2beaa","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/environment.json","sha256":"9c16a877b83eb2f0517dc30c963a59861863b645d8b52264bedbe9e4a3cbc3f2","bytes":5171}},"requestSha256":"e44a1d58fdf18d5123638577599cd5e11429ec9ffb6c3de338cd5c9a9bf3b719","metricsSha256":"bf9985f96ac65820b262caf2d920f9cbeaa3afe1721b8487295f579ef3db1635","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2452072","recipe_identity":"1f44e2d5a3a3a75cf74db9e8248bc6d0eaeb94c506f7f67214683896c8210e9c","loss":3.577184415113812,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"09a1127653dcae1d6b7e0390189caf08486922c45fe6aed22455b1e5f65e5960","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"55554f8a32c075ae3c571c8cf41526c3f0fff476f5ac32ccc3b4aeb0931357ee","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"9ed21cdc7f611ca046611e1de070ff093ef44f7e668c56923aaf68288feeba79","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"2267310499ca90af2a97b51deae5f71c9c3dafb68ef48c37429e806071e44dbc","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/environment.json","sha256":"5b581cc10ac5a4dced494cf4a355aab4c976e0a1f06c2e8e1bab25c0d49690a7","bytes":5202}},"requestSha256":"67040157749a833bd52c3e5b56f456162b267577640c0177d90459e72dae37d5","metricsSha256":"9bc38ec9978be15d6a4fea6310471e6bca0f3f855e074af10f5e875c14e1519a","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2452420","recipe_identity":"1d4fbaee29ccc1cb6b7e860619c5e8d1ae2082224c2e7e84ac489ca86eeae96e","loss":3.571636840380292,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"688ab843162e70a8d400b7c07e9523c6cf3d2701274f92e710e3c30bd0fd4a67","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"10b04569575c1960ff57f568d8f27101bd34205577b1c294cb2845e58e916c7d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"56154f0aef7c266f522c4e7c6f335c5503e167d2d88d797301e0bee844ecf3a1","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"20b0cf26dd130f18f7af78c5a942ae0d393138c84b216d7a3e568c1100ab2388","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/environment.json","sha256":"a614d0202911f86efcefe20b42b17a59f73b85f3d53c18301ddb968b00550087","bytes":5197}},"requestSha256":"ed28cab1bd1efe06669e237483c65fa7a7d5c6afe0c675b54b62401ffd82f343","metricsSha256":"4600bafddfd3c36559cbb63a5ff662778234e852ff8192ce66d837a8764c7ebc","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2452421","recipe_identity":"f60d722685c2a4866dff4e16586845a0b55cb1a00e1ea8e3a4f5fe2826515360","loss":3.56448257767416,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"be13145a629b67ae74702e39bc95f937332a5b2df12064a8058bafe04b4adebe","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"4483f947f4d3aa718ecd1b0dcd377189aa103494a07e96bfb8b46120de64de58","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"7a649c6661901ebaacac7160b05b0ea0202c74d7b5d22de699320f09e54b92fa","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"45d0a772331137dad76dab88ffa682da8235831549139ecdb6011d49b191138c","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/environment.json","sha256":"4f56b32de58742dad1bf7fa4cb0172be494b82ac704981470f78e3f17e9c383d","bytes":5195}},"requestSha256":"9f77662083c30f339fd74b7776ba261d402ee9716b651c1bfada81cd0f57ccdf","metricsSha256":"1671274d5756b2b003e2177fd47df56c33f76fd8bdae5abbcabf1a7130aadcd6","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2452422","recipe_identity":"8c1e550e2f53d514b1088f13b06292b79eb7b5384915a73b24eef0db2d2abb40","loss":3.5674018363182567,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"bddc99c82a5fc616ec54e080a99d14071830668814349dee240a3a1d0ce869b7","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"c1971cf4330d082a75d519ffe375cea1dfb103e5cc6e00855671f41b326ad054","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"c1512a686e2f9b198a4c4b22016577180dfefebe35eb98c9fa21d68250e97133","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"75e3163d875d1f906607628cb06ef60f409138bd626fd4835ac59a4aba6109cf","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/environment.json","sha256":"70ce1394adad163a87c3d1ca2b78dd219889484f8dfab0256d8ff43e29ffbd18","bytes":5195}},"requestSha256":"b5f5fbe8865de865dee4d7c51b92c0ab8d26a6e109797f95bf00f705a8ccebe9","metricsSha256":"e999e0b31027a1ad01891551e05bd5d58f76d61724aefcf80e1c70ed771c6e56","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2452423","recipe_identity":"eeab317f65d1d74056a9937d793fb495a528032fad1ec3126d2a18e40a7059f2","loss":3.570829121158611,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"b0a8c42a5b4ac3299ba6cd99269c25228a60b095aab1c1117e5407f06413be2b","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"31940aa28b2b88140b041469c8be09f698db96cb54b9f2924c1e6aecddb8b242","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"76a0e7b69209ae8ea1b6ab93f601d9b9f7a5ffecdb9710b214f1a26c7e97e105","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"a58221a4cbae212234f50a9c566b5235e1d3fda80993dd0668f6fee002bfbde4","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/environment.json","sha256":"d9e00b600ba8a6841b32ab5386c8ffc3bbdb1075b5a95b553cf4db9e7ef838e9","bytes":5197}},"requestSha256":"9f0eb92ad553cb74fca9fb4c8ea4d50fc55b779712a315a95016222801101214","metricsSha256":"0d08177b2aa81b094b0a721e9907a02de0be512bac5eae6919cace1fc83e4bd4","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2452424","recipe_identity":"5dd5fc2cc428fdf8761089f84a14dee1b29e32069a6700a9b190b58bd610b72f","loss":3.565451243638595,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"db325328cf51dbedc4ae34130bcfb861056ac4e0eceba48eeb03a1c1a2d87329","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"f464da5a245b68d66919873333b3881e06fc8c442249a2280b27e0d471edc7f0","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"26678bf2a92cfa2d7370f6162bf4a054ef7a7985d71f7851b7bb7ab08727d946","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"371af8a05f2d85aaf49336f855698d5f05159f078e916a91f236bb9d3e62d85a","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment.json","sha256":"0a329b9b8df42f9c10ff2a31186b13553ca282924ce5aad70f6465120fde0695","bytes":5195}},"requestSha256":"b14ab92de08b4ff473ba2610a786e01cd80a1f125981dc6f1994ede2e5d7b428","metricsSha256":"2309051ec247aec3d2f22e26f6f857300dfa5b4a39c5a96647def057e65801a3","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2452426","recipe_identity":"584e34cc76aa15b70badff5da687df642cfc1762cf9ee0253038378e5df887ba","loss":3.5748634036493634,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"566b02970160ce5716937b2ef452df8169557be16b263f220b66e22b6fdcb775","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"3feeab752ec5df3a9816555dc51e411abdad092e3bbbd5e30ca0c0bd89577cbc","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"a53d98178e7fffefbc8a3b088853004d76793ac08effe5293935f678772f1469","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"9bafe8919ba2a58e72e7f24bb94a685c9d8b1be4332cdbef2e3a70bed8b86a3b","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment.json","sha256":"2febdccfc72b453d7e904b9e3793a4189bc8ef4ae55e432b40870572d3b318e1","bytes":5196}},"requestSha256":"f247c5497acc5059ba04ac9bc99ca1291770921931a798f49082f80a00300405","metricsSha256":"e6f5da8cba2d3646558b9d4f0552b9fcd5a08e909736ba6936c0992938097e06","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2452427","recipe_identity":"b516738123645649eb78cc3175378ac7fbabd40c5a9bb118a3799d50946a217f","loss":3.5736172870777962,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"333fc032e2286e7b9aedc9c9762589c04add13482ebef0d59f72c6fa8ed4fed5","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"8ed71bb284a5efd58db5e323846c9041bb05cb7f85daf7749646198bd9022fdd","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"40d438d704a55f0be54400b0cc8460e5e30b0f59508f9b1d3abe675d8cdd6e7d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"b7abd2c081ec2518790ed677132430d53063c94fdb02c4a3b432db7d68416a32","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment.json","sha256":"462bd50c483855d02984a60cb8195250eb343639c2a0b970e16d6616bb940577","bytes":5168}},"requestSha256":"8a858748eef6c8ff93b506a4eefce1fefb50e457619a582ae7218c8b23b5d532","metricsSha256":"c0d25481f80195cf275dd1b024bb7905759c83f8b2da3cfae0cf2f84d608bbab","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2452428","recipe_identity":"bbc5110848251cc53f8993ea9be4ce46edfd927e7ccdee4cc9e1fc84aefc11c1","loss":3.5702033761528265,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"cde998aaa62f9b9d9c940fabb76f96ebb299229e507deda00881b0d743eb970b","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"e75c9b6a6bc753455d0700c100b443dd74174cda9838bc223cd01ba680973649","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"89cdf7cedb7f6a118c10a85be691f8675b586a0ba177a25281bcc282fbaf4b56","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"73c2415cd5ddb6b4019ccf5d849711d3458658f0113111b74c1fcbef2ec51cb9","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment.json","sha256":"57a9f0cd8596a6614fe4ebf02b658f1629e1c2d11430b60cbe89758b7b968f92","bytes":5197}},"requestSha256":"a7160c4a499cc145959d272d7b9f1e7f6a802ec61ea8a5bebd6c10cd503bcc56","metricsSha256":"4ee1912eb242ff201b70d1b4bc2db547194dc76669a5e522d6af3783c76d0e09","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2452429","recipe_identity":"d0cccdc814580629bc2ec3b69fa2fe573ec3c07cbd04d49a5b0cbd56917c4d22","loss":3.577544288226398,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"5c5c81ceee0c8692adab37013452c1188a811fd702af0eb7b649fac5306e3581","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"31c5f8bfa494487f466f2aa87457304ae3b3a080e5d000c0bb67e1f705094b2c","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"8233bbcb5f15d3a8af937a75af6169d89c06e2290d27266135110f656b45514f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"0118546d94eaa02101a4cb5353bde0c67b8812902451ad53b495179cbf65da21","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment.json","sha256":"fdb4560783f75ba6232145df18fbe5a45f635c66bca630a4c9a15a7986998d06","bytes":5168}},"requestSha256":"d701d782bba1bcc7bd31e01a622d0b15f3100fc58230210ef7f85bb2942783b8","metricsSha256":"c18eecc304f720e7a4c868aa1c3efcd914d53cf82f7f5831496ef2f6c8a370c3","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2452430","recipe_identity":"c7a99c746d212c53e6e2dc86c00f3a44d116b13047529f88d2671db41da0adc0","loss":3.580599891553157,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"dd87df98763eb25fdcd426225366c88dffe59bded6f8c4ef8cf21f72eb57d133","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"7feaa4674d06d3472f3218648b93a00f5d1bfb8213c98ed0a5a04ae4c5138600","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"1c545ac7d2ae3053a03e2b8270aa6aa087ab4e56825b72792cee558612ace40f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"4e8116845855962cba2573d53c696bb8f62dedf892d1112d7a702b336aeb6d98","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment.json","sha256":"5ce2ef9c40510cd173b1b96e2a5f5dfaaf336b65b8f274a228d32e58f9e6cf96","bytes":5197}},"requestSha256":"5477c242023137ae14887b82b98ce5cf270e44c1b147efbca567eb4588a428b4","metricsSha256":"2ff53c6770907511eaec2fa39181cc5c4928d09a82d7a911ff588e6f7475e010","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2452431","recipe_identity":"a4cd01732ded3495fc208c7edb0aae9386e7699d42e181edc6d60f4b55d01841","loss":3.567332756965113,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/result.json","sha256":"15a0ba98ec8991f38a7e4c48845aba6c908c56c71041845b8661b2a439a13130","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/final.pt","sha256":"75d3ecea745dceeca15016909d9990658331f8b3e93ed3ea7efc782fee91a436","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/final.safetensors","sha256":"4b02490995131c2ec9cd011f919c5cc8cca2b2763e2fcac236ab9c8385de09a4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"69ac377577414e1be580880bd7efeed77f05e81c3d1261e136d29c6da8b91fdf","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/environment.json","sha256":"77abd921a11319bf723605ae94b9dcf0ac6ed002bb4175790a72d3d5f8388db9","bytes":5167}},"requestSha256":"5d1f92640f82d22477a38f37203b729f6d8578380ec6dee96e1a7876a83fe07b","metricsSha256":"1f51af252c8e0d1160c793672a2d5c913262c9889feeb9ef61e11096683f2e4f","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2452432","recipe_identity":"dd1222ca36bb2aec2958e14601bdec6fe636f23385c464c49e655f64702319f7","loss":3.5687355655742645,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/result.json","sha256":"6f56344a909341cb08d3bc3c4f7a4668b223144db8069acd2a4a63eec303d024","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/final.pt","sha256":"b6da3820483f4a1a60db7d4fef889e55982d4e1470f63c001a44140737c3444d","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/final.safetensors","sha256":"58660d361f1e368b08f248d09ace08e38ac4e1121eb3229901cbdf3f1f6435ab","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"a1aec47cff600b6fcec2968230e749237a84004a6fdd666e1f7de73a36667f2e","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/environment.json","sha256":"8fec8b0ee5730207872359b35a42529b15a1d044d02d3adcb4c2d7e2ffa6e5b9","bytes":5196}},"requestSha256":"25ccdf4e86991a1e75d27d597dbdd9ac651d0569b73e71ec8597df65528b62f5","metricsSha256":"349bb0ee1a951e447893c22bd0a725a2ded8d19347cb5f6b259de305f032c974","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2452433","recipe_identity":"639452fd324355f5e402d15f0d912abe67c58a4966afd642ade77301c70216cc","loss":3.5693477999073466,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/result.json","sha256":"b1d1c44220cc695afba550b95ba4ae9ad03fdf5b95b23bd46b19cda5eb1d52e2","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/final.pt","sha256":"3a6b1fee9f3215b89eb94e8ea1866170f174ed6de3daf265f038ccd5530e0a1a","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/final.safetensors","sha256":"ca8f38bda6a46cdd568e1b6c207b5925e6af6450a2945df9ac7b607dfef8f653","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"344481d0558f0fc2c26a329c9ab510d551eafa59e2a43be9dc9d5b4c519747c6","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/environment.json","sha256":"60113274e6af1d76f45823cf69fa532894242e4536e2cc4f398ce58b9e934365","bytes":5166}},"requestSha256":"ce998db36db291ae1529996baccf13dc118d5a8d161bfca206871fc80ae2199b","metricsSha256":"0f0395e1c0fe7dc88b318c6661436c300f31cd5cbf66cca513fc6e0fc940e224","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2452434","recipe_identity":"eb8cf61827cce2d9a57acee0a4ae16026112d8bb321d19efa7e1681c91c243ba","loss":3.559489830036192,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"5761f604ebf797d12b8a78e8a27315cb681acb5d41212b920e95e7661b37bebd","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"6931de14b8dc6d02fc4fd1048a5bb0abc623e84e544f5ae65740321546648243","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"a6ac4c5da33cc3e88efc406310e969bd8ba862225a0c14dc44ea4fe316d6b73b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"ceb96932cb8ccb7f0d4eb3d37f6f0445d24f1c31542b1b5485287dd65018846d","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/environment.json","sha256":"6e5d1201ed0f2dd1f2a29c748ef65852b66259999b0b1fa6944ccc29506be1a8","bytes":5108}},"requestSha256":"bbad30cb70f684de9fbd0f1615c1664241aea67016e975706bcf07cad295180f","metricsSha256":"152b1ed1860ef89549558dbe399e7ee4c27d092aaf0ff998819066c66780b551","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2452439","recipe_identity":"599f48ff3fffe43f8e1f0c560f21db5fae0a785d24e31cb0328d6de34983b376","loss":3.564557421940719,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"a205f09b8fa983128c753d777ed1b2c6ca36408f99da3d695e610dbcc06a95cf","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"49a141318016225b6c231c83b1ab88edbd05fa668f852241758c9692ba20751e","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"d7f57a7ce6016963155196d498a169a3f01c7bb808e46d7c565840ff7c3efab9","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"11664c7a8191b5b05c1da361c2623c24eeb8e4a453998d94687ecc8e2ce9a7d0","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/environment.json","sha256":"87204a3a18bab7e311e2f99d2ed76035a37fac95b52628c956d5b518c9d70eb8","bytes":5194}},"requestSha256":"dabfcb12908cce71803a6732e9d7a31f5428e5ea1f80dc2235319807afc5300f","metricsSha256":"b2abd0b5b22f0ca9d7304e7c1423c2ac4abe75730ce7a71c91a3bd675d0c734b","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.006-s44","horizon":20,"attempt":2,"job_id":"2454799","recipe_identity":"6d450712e6ab215c4d8e711df72b90edba1d801e2f668aadeb9da64a15634512","loss":3.565673217273487,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/result.json","sha256":"3d45cc8a68b5f23cb9cc87541e2150a7d3644063e5fb588715af5a3699ac320b","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/final.pt","sha256":"3561fbdaf83f0167ce097e745d64860a6668b94b16bb25c36c9d8215b0538d0c","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/final.safetensors","sha256":"a105489d76860b4cabb7155b41063f71227455a522d8e5b99ff50bb2c3e57d85","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/epoch-035.pt","sha256":"0da36d53298b57b321d53d9c5e97527aa04e5e8e21e4c47368a13de2f730f34f","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-002/environment.json","sha256":"a64ee84edff8b65893bc290968ef6348761518952b222a3389f3675a14e487c9","bytes":5196}},"requestSha256":"75741594f8a184b5ac332eba4e87c7085554f4373d78ae0e651086c5e228cf71","metricsSha256":"cc86f1519050f5c7475ba80a6531a52ec2ed712e26d86292176614db75216707","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2452443","recipe_identity":"b9917fd47aa835a95becb81e8b319d1714035d318f9ce084d668252c8390a7e7","loss":3.5544962947525507,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"6276937961dcc1efc0bdee9544a1bfe0f31dcf9e7264e49dfe079b97cff568ae","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"d4dcf4c900af22be4b0e26ee82aca96404a4ea09fb365c3af6ec7ff95003ecab","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"a9e7258d9c38da1266f5c9b9f623ca6c5f4e645686f41f4ce9248480f150e0c9","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"7de1ad519e9204542e6688b6f36d5e614375a226d09e0a46edbed22ad15391c3","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/environment.json","sha256":"610442260763eed2eb7c9db782a3a2fc719e5c239c414dd9b81d864bb609266f","bytes":5171}},"requestSha256":"e7ca471780aa7f918fc5198c5952cb1abc5f12a2c9b5205cee7761a7db4a0c5e","metricsSha256":"58ad939cba43b47f556af575d1ea96d8ef4b2300340c73b626e4c30304931b3d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2452447","recipe_identity":"d7d0f17990cee0856327fd6bf5ab3e82806bfbaedcb5ae4040c33a1a701dd38e","loss":3.557062698977349,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"06c9b69774f390333422772581e07af868bc4a19da893b4bc1a1de33784a6f14","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"e5d0f09c4b1f7ef3cbbc81f8d62efe745b48bc319326acaf08dec8416d9d3d22","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"6074b644831cef098eb22267cbfae9e4ecf4fbec9b125611fee8e2cd6f7ea99c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"dd18be177db94f1fdb61c3b229bc83b40a164ae7ccee92ea3d4c9d96a5808d3d","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/environment.json","sha256":"40322af685cc0304251f5262ebbc5f5f6c7557c4565a008b634fafbe12f9a4d5","bytes":5202}},"requestSha256":"cd6c56b0e24f4c08f0a5a51aca89f4c574b972c41af0bab1bff1753dce630e04","metricsSha256":"2ae0964fb0a9d2f04830885d95abb750d1abb280dd981c3486a1344fc3bd7148","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2452449","recipe_identity":"f47d72255d99d1c063cd9af85acc3fdd7f7b5ef6c4f5d626bc5928d0463506f6","loss":3.5585373031785137,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"5445cd31cc5809965269d1b81b85f31200c88973d344d7ad1f6fd6e7c69b0648","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"9628d387e66ae9b6c47eea7d832229d3f0df03dc30635ae75ebbbdbbf7d8c5db","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"3655069baee8075db9a4acdfbbed6bd5c9ef7c9f3a673fa7bb243ac305b4fb6e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"ee1d7e6d43530e9d6562a408ea8c3003ad6d514b14913843fcbada4a41d03119","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/environment.json","sha256":"835138dfb368f1fb659da93e303639954f4c8f0ab7cf39938331ca968e704cbc","bytes":5201}},"requestSha256":"730f603f8cff0983444e75278360142cd8bba87718fd87301c103e40d093913e","metricsSha256":"54ff8d1495e1a7222571ac0babce4a9c85fb345a0d7d4f5493a88de95483ec3e","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2452450","recipe_identity":"3806f8012353f95b959acf9f17d705bbb4d0d7d80389974fc4fbffb9377f2514","loss":3.558380087390087,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"99be1be84a325c006be5539bd53a3f5af43a89b315bea07cc2be81c6ab8d223b","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"1424934edf422f8469188782e43f20c8cc66edad8d29029ace1c5a01fedd96fe","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"d7d57b204553e12edf16bba038917c0624218e5b1164bee0d0562f8c5b007acb","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"ca9aed8d3076ded3de2f70a98530db3a4c92b8eec5203085f49196a08f48fe4d","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment.json","sha256":"5ff96788ead156f48070182647e128d547aa231f49217dbcc82f7a2c9f93ab4a","bytes":5201}},"requestSha256":"741e7932ee62113547b4d3d172b20182f09ccdefc9b17f6af8c1672923c1bf5d","metricsSha256":"3861d799be7e0974161cc8f422c0598dbe035b4b455b0d6fb609debcad965cb9","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2452451","recipe_identity":"bc727f09b952ac9ff11ecda9173bde1fb5d21cde47597b56424071a8e05ea15f","loss":3.560115019637252,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"60361c2249b8189deb2a9a6f2464f95f35dee22cfe59103f2804e71ed460e15f","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"654f244ebeb0f8a124b3d5595b32ff5c7c64ff9a6d87745f144020790c01626c","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"1183eb9e0f3b2c59b0fe3a3a293e828d31ffaf33085e126c3d5b434f6de52175","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"a695d78f11f98e7ac1aa0ebb50b23491db28a48fa270bc0841e86390927882d5","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment.json","sha256":"ecc82b82f04a285c9d34b5019a0d0397b7931c280fa6c3c3ca53179a39372564","bytes":5115}},"requestSha256":"b0db4b1fa9db1d31516bc448530bf7279b6c4bf780fb07ce08c80c6a75767e7b","metricsSha256":"6997bd4ea0c0c2f0b25441d7758c05c84dda0adda155fccedcd8b615d7bcf5e5","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2452452","recipe_identity":"5684cc969d18e7a8638015b7bc30cf72355ba58c28cc373c0ab30844c5bb298a","loss":3.563765453203556,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"fb9249d6bce7794627fae0c68ec016978d223194c09d99af3984f79aef1af192","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"2774334fd05be457666e33ce11413e588ba805920ea11af4926ccc888ef6d172","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"16f478781ecac58c09bf6f7a9c6a316c8aad020789fc07318cbf7d018caa63d3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"6e13e814ede8ca438628d45b2251ae7b3ffed619a8c6c04c0f53663250ed1cde","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment.json","sha256":"d5b18c3e3e922786ba5757e046114bc07574e2330e24107c73f288c100ed00ab","bytes":5171}},"requestSha256":"671de943c93a91180058f532384e2d45d735a96814ad504d887c95a3631dcb1d","metricsSha256":"9ec142fdb1ebe8cbfaa1d20de510bcd7fcd4c4036c3ac8362d2aa6f76ae8ce5e","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2452453","recipe_identity":"00f7c46b70aab600be222a583e855cbf1235ab9f33b0d3e7a1ca4ab2957723b2","loss":3.567208347256614,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"174bc072a7fa1604af5f1b5bd0a505a70730e94c67587dab7f2538a5c6a7ac7b","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"e0eb243a6b92335757df034e16ce10e9f3516a5081873ca858907c1e8f7336d0","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"92cca95a7697169ad9f780bfa53c6c73e2a6dfeec18aab917eaae42e9089ec01","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"fa8a5cc637da059f088a96704ffc37b7a57e40e1ad7ef4bf209aade58bdbea33","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment.json","sha256":"d9cb1b1b21b937b92c14a0464f4edbb1521c5cd28fb28257cc35597bda922c4b","bytes":5200}},"requestSha256":"b67d432eda5c4422673873c1576f14f53c008b743ca3152a1485cc91eb99e32a","metricsSha256":"fa9b125f8a0b86277548575b87938b552b84fe0e078d797978c34f47ae864723","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2452454","recipe_identity":"1ffa38904f628ed9d48d0eecc32086871441d0c2609b959096f5cd2bfd206838","loss":3.576001709655114,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"0ca2dcc48e453202db144a8d62b8ab6563e5f130ce13b202cb9b810565e925db","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"94b9e192c1efb8a6970642e924191d7147d94288761d4be00a6d314b16924890","bytes":244949995},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"4a28a4cfc997649416277ee1f9083797a3c0b6dd27a492770685424b045d8638","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"f803514e5f692e5f477351a7200146210deb428cec7e4d2804ce3331e77f77c9","bytes":244949995},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment.json","sha256":"8882ae9b30f89c959b6423a76e8d90ae49282edaf483f4aa4c5c38f47f8bcb64","bytes":5200}},"requestSha256":"01290a1c94aac50eeb9004435991e1a76856b06828a903d7a7591ba067349e03","metricsSha256":"d5e42732a07a58a3dd23a771a88269a4a8c8806b352cd198e9c14489b7c06a9c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2452455","recipe_identity":"916bf5b877d6d81347b1922c7a4c7180398da5f1bc34b4a7acd01dfa50a4f5ee","loss":3.5661558365932238,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"38bbb14c62715920eb8dd8ccd3e211ee47db40a0412a9e374890c6a5196497c7","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"93d2fc8a07662cacf9a97c89efd6b7b869ff9fcf6b1b7cc57491f4e0bf9a24ad","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"fc67f7d3bbedd4e66aebd31b71ca02d93e626902019f69ef3a847087a1214b1b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"b0b68a407bea7816f6b675b474528269734b1dac72e00f6344e67d27cf66f56d","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment.json","sha256":"307b1879eef529ef440cba2fa1d01e907378fcf8cbc1e62769d9a25494917e94","bytes":5114}},"requestSha256":"524d22449fd83621f56b716b5d63aff211e0182df8766c976d9b35846e5f1319","metricsSha256":"a992418675ef468ed1132824b4e1ef62c436606fce9d6f05dac7b8049919f417","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2452456","recipe_identity":"02b026372048ff204e10fc309c8c53e3b64c8f47d4dc828b6f93bb238d6bdfd6","loss":3.6366870581997817,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/result.json","sha256":"dbea7b51309b3607d8bae06fa6707c653290260209229437f801f5f85e36e9ab","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/final.pt","sha256":"96db371c353bafaba4d3dd8bd3719ea71117b9ccaca5c44a54cf923151c491a5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/final.safetensors","sha256":"df8b4a5dabcd47ed817ae4930d3bc5162ac1db891ba3f13c460f3e9fbff6f197","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"fe4740974741d89d68c45ff878a84aa1be0e8819d8333663dd593d153c551334","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/environment.json","sha256":"75278b3d0dfc11f7ed1ddb4415f4baa99f9de38f9d36783b81b2c9b38adc632d","bytes":5171}},"requestSha256":"95d4294b9b7d081b27a4b54d1ca1af086f092c987957d9b2974f789aa12e24c3","metricsSha256":"8b844f0d575fbe70d9128bae8a5fbeedc5079d517144f169c499bee0e10ee14d","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2452457","recipe_identity":"1d95394fa5d954dce4cbd6e7dadd6e7af12122c7a38fe96273859c92071b4077","loss":3.592016682530186,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/result.json","sha256":"369037b5b65c30fc780c3223f32a5f553760801801ce99fdd71c2fef82fe974c","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/final.pt","sha256":"22ff8373667e628f6d863bb6c534ebfbe8eb1be43014ac024772db87b50d15b4","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/final.safetensors","sha256":"5c12734201f685a330849f5cf1b5867cb5775c733805f60f570a01e48f1c0874","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"465c03ee39049b7d6bab4f89e0a3bf2288022a75a5ef82b29eab1fc288df1848","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/environment.json","sha256":"5747f71a98cdabc8037f880f10b84863d6e5939a1cc5a96efc1dda2d95d1231d","bytes":5173}},"requestSha256":"d99dbb7821c7af736a44deee9d12588deb47740a60451f19f960f36af7a2e860","metricsSha256":"4b9f96b440e4cecce80cdb52c5516a27a7c20239e9b8b3ee33306dc8cabc3d92","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.008--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2452458","recipe_identity":"015e5851ae29a63d5e14b64ba722045f84297d4f82735adb9a52ee280e68395f","loss":3.583236902938327,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/result.json","sha256":"3663b75dba61b7dd0d7825d9f1a46d1293bb83abd00ba1ab76f3174bf6ac7f82","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/final.pt","sha256":"cbbe16e079dd47e0bc042afa662e2bf520839d5cb8cdb61253bac34ba56236e3","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/final.safetensors","sha256":"512145258918625fe9e3d2c97775196b7373d453912de8ef61502da533ac6767","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"da67fa856d924b4a98e80e31273bd7650572af9c8c9b853cb414b5ad4cb5b8fc","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/environment.json","sha256":"2b9a76b026aa3041a9c93b265e1b974427d7cdf5cfaf653b4da75389a17ec040","bytes":5115}},"requestSha256":"26b32436cf972972b6efb0d5b824c35ef14e146942da0638741ffe5db42de7e7","metricsSha256":"a6f6242623957738cdb1f3730ed0ffc9caf80ffe07931de7ab6b64341c346ddf","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2452459","recipe_identity":"878e08fa30a42fae12d1d483c88093b533381fbd5cc38c52f68681f0e9d2b5b8","loss":3.5781384521506574,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"997618caa623ce446cd8201d172ac612c0107cd919638b61daca966aa848328b","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"1a3aee744b5869c828904f9916408a20fd54b849ea23b306237d4512fd085f6d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"a4281b09f3f2e9e03cef99765379e6b12fb899ce6d09f8417993339960f5d44f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"7e08588a9bbcbfde096d9cb72962ff5b7ecb0ee5604321ca28f5453700c0bc2b","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/environment.json","sha256":"948a220f0c2e9003d28bb9d8fcbb45a546f0ceed61c3126429529549a51877c5","bytes":5200}},"requestSha256":"ad080573f07f9ef2b554d7796ab4e5df38b53e93f9dfd31b376a80efa612c858","metricsSha256":"70c8dcb12082c5917fdb792976253a196a027ef6e005b382af2f978dc95118ad","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2452460","recipe_identity":"46a3f8b4ddeb3481a2c7b935695194ffe84798d081e948e2916aad37226d0a79","loss":3.579306858623379,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"f30eee461e8c5762fc6b4e36a35b95a977fdddd9be2780d4ca72cb290a63e069","bytes":740},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"a8f80301a1151b7837fca2c46849da41bc6610a93a485daae28c32e21de5f54e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"567f774321062419a041ffb4129aa145929830fed848c01fb2a220ced7371c22","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"20ad166a8951a51286549e9a976618fece450c7c3e7559407af577641216e5e8","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/environment.json","sha256":"d06bd1560db73e9bdf9bd7e9724914b5e0b4375ca358fad7cc3458cdade6980b","bytes":5116}},"requestSha256":"25f539f69445adc7b005809adee0d2388c2df2bec98e691e83975475957d5f77","metricsSha256":"57c0ca2b3901d526d61b8afa729f8df7004f9400c6ac1e366b71b8c2a73c058c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.006--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2452461","recipe_identity":"185dc45642683098ce2e7cf393a232b78825a31abcea2fcb5829b43b1605e158","loss":3.5734827982720696,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"b22363e77ea90987f7e112cfb1d5212422aeb93493d7bf9dce8fb723ee543cae","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"57167905f961b7586b7a2f4e631ebd4dd9d656b395c60c343ce7488bfbfbf2a8","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"a42e57abaf733aac6ffd1289bcb3b3522d0717bf2dbfed3e528e4486467fb451","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"9eebc682ca1ccafcfe3a148fa3b959cb77092fb87c0777d9e97c080f15f5ec86","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/environment.json","sha256":"57d2b80b337846a745c889072b26fe3d92cfde123f6e8c0a6b200f07b4ef3f7f","bytes":5173}},"requestSha256":"b3eaaee5aac3a10de1c38b9fc7010944e55356d6107d0860cda1baf2fb5b8c55","metricsSha256":"5c72d53af9737526533587b924467af141569ccd48b10e5182f4cfa1b2f4c435","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2452462","recipe_identity":"7a9def965aa676ddaebc4af68bfc07c3d891886cc1a5220273a310e1ee41a002","loss":3.5627517779599316,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"58e1a1fab73b52e2e49fa08b8ebc13a0ddf79e3aec3df8c24ed086ea737d9dd5","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"cec0322ee41597c2d56a28bed6b1e0b8924ad18e06a0286046b01924df8dcaa0","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"27be80103344e37d7a332cf13ef1825b9c34bf6bdd13679816747e9e53b5ddc0","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"3ba0776950f5d73be0b3d3a79097c5282c683fbe246dc74b2acb9c605132f40e","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/environment.json","sha256":"d1e8e90761be2c8d970fecca75b3e862233a913522414c299538a5265e1c1743","bytes":5202}},"requestSha256":"9efbaf8a7500658bc5aaa1fe21f09e87d9f737752c79ec0cda2b0024e544480a","metricsSha256":"c57dc5e70dc69327619bd4123c554fe2e4c09c30eef74cb65c638cad8a148ea6","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2452463","recipe_identity":"1839067f3a053254a27d3fe93aebaaa467b0c97b8ba7c19bc95d00f72311489b","loss":3.568954754386764,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"9e82e695165bdc034185f0938234ad039d64dc773b35c9f865aaaf678c470355","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"dda054c95c50a70829b75de5d972b4864214b3bab23b5249c3da076a231678ce","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"67969e83bcc3815fa75d5c8f54c847fb7b84ec96365523d67150be05f43d15fd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"20eb80f0605eed7a2290120d04fb40c15413796ec0ed75ac129adac03dd4e6a2","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/environment.json","sha256":"ad71745fe207792c98aaaf229a662148219d4b21bf295e64f253350c1009d783","bytes":5202}},"requestSha256":"8827ecec4d6e951b6a8bee8c7454a0bef500d59b0df8b66a77044c5eeca544ce","metricsSha256":"ca059e295aaa8e0323fa807e167e17db71568ec1e2a07f1966b0fca041581926","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.004--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2452465","recipe_identity":"012b311a4fbb439c0aa5f9383dc99ceed882ddc80e03145aa2aac5b23f4ce6ce","loss":3.5696288155343874,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"001f38a2a412423d72d5a6b8375cb79beb007acbcd1b54c68086a816aa47c0e9","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"ed86fd502e135f9755f6dc9e568d86997b65f2897989ebc341224822072bcf93","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"56a584d604b4a943e01be6a900ab18e158ee45637823923ac0e4fef75fc7d7d5","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"9cd1f82ddf28ac0fbd24686bc9ada19a2556926979a463068cfbcbfda7992fdf","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/environment.json","sha256":"3250a7fa4090b9b571bcca01cba9c5c4858ad892f1f1b83dd4e90fe7f73b9928","bytes":5202}},"requestSha256":"c6d6adec874ce396c04b2b1b073392a8ab42cc8f5175fb2e73868b12cc9a4bcb","metricsSha256":"728d799a5b1a16667e8d0012879a6d8bd9369e9fd69da0b5ad7ab8e73a896c54","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2452466","recipe_identity":"32b8adddfc7344bb26b90362c7c9a892c4e4270380db2e0f83469989e35e9d76","loss":3.56926195266135,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"e8dbf16cc1081f0fd5f1bd36e519e03cd8b18205b0331b4dd83a791b119315ad","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"dc63da4784c96b569fa7293732fd06239fb7b79833766f3ac05f004b19f50c98","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"ef3f6fb0cbce445823377e254451bd76bc0d66f1ba13a8130c78eaf9fa043ea3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"99119f3f75eda94fc83663232ec673c93f5037986984ebf514cc2d640af65e77","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment.json","sha256":"5d9b30140735c51b3d14e19e45c812f483fbbc665cc15434dd47b27f44ffcab8","bytes":5113}},"requestSha256":"03a8636243f10111cff965def5046095a5f286273d9d577b369a22a8e3d8300b","metricsSha256":"e9ce2aeb927489e4025ef272332adea0df890b220a07de18a0008530c5c73140","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2452467","recipe_identity":"6472bcff2340dab0523131708cf3d0caed47268831ab992701149e32eb39dad4","loss":3.5674922862088256,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"6d80bee295fc8527e7420849c66beb688d76c590c50918cd7d9a6af7e52493f3","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"4a996fd08f28057b429dad81d64edc0a27879ca64dbd4149a1382ae192941949","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"a2de448ac1ed587fec2e543c53c5e59fb932cb90e7895b678643b5baa675dcec","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"ad2652a0c097e7843d0639686d200693e7d2f190f2c64959c90eff209e602062","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment.json","sha256":"fb55cf41214a7ecaef0d059bc3afe26ac8c8bb3ebc7009b17dae0300964cdc7f","bytes":5200}},"requestSha256":"8ca578fe1911b8e7e965c5f353649cabcc08eb99b5449547e47c4a59744a5d9c","metricsSha256":"ec17cc22e8cd882ece404b1a67c42162ed35587f2aff79f4f31701f98a164baa","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2452468","recipe_identity":"dee49cf1f9785a4bf1d721580686ac9814141590e8282877ca95703b3dd8842b","loss":3.570816793700776,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"5f2ca04e3a6edee783121f00fa50804511ae9cf8fc7925e09766107f130e5068","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"d35cf1c886935cb3e054adf94b91afc37cb4dcbf5a6612798c98d51b7152bd55","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"08ae696b7ebf45ef6e362005fcaad8057ab67a093e5ff39b1276191472e6a484","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"0951b0e0cb6bf38741548208c74c10df7d7d7a066a742948e52b0a674b7b8de5","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment.json","sha256":"1711fe1647f22227a61fc88656e947091bba03187640c040a1eedf20f3649f59","bytes":5173}},"requestSha256":"2e3a7e468e6371e8e69ea69ef06a172cb5006eb80be5aa5f3d3694a7ad9578e4","metricsSha256":"5279f3ed4a1dc367cc2e8a66b5fed6b3f3f90a66f6cf530a940130da733d007c","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2452469","recipe_identity":"9faa0f5892a72d47793cfc6cdf09077681c9d681ef1b4f0717930aa2a1518b49","loss":3.5726546033636044,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"fdb36a468bda1f5bbb8f7521213752029077ce2b248f255ecea284b5a4f127f5","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"51372ce24ee1643cf1eb0753177dfeafe96d6118700adddd82e83ce55e0e1b81","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"3994971161146e3795f819101d424493b80c8b7a5e08ac6b1d9051de0fd6a23a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"89f1b1e6418fcfd72d0e6be546ad0eab7d376b1c86f3897a0ca9087380e42929","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment.json","sha256":"76e054b18ace45511ec30fb167367cec3f48723980c58c4f5a08d3edc717d5c4","bytes":5202}},"requestSha256":"59fdf9ec29c9a9bdca180fe30d526604ed99aa770a92c0595e056584af57ffa5","metricsSha256":"738d2c6a872b4dd2784d48459c197d1cc6a0aad3c38407ba2908073cd6649be0","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":20,"attempt":2,"job_id":"2454800","recipe_identity":"959f20efd4d07b81c4ff1d15147b0466e5d7457c110e7d9cdb764d88703367d3","loss":3.578716270790343,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/result.json","sha256":"8d14d743985f863d7ae31f36ec4f0457dde60ee4ef1699d8dbbb38579e7ffaa2","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/final.pt","sha256":"998905a1601b5c0cdd60cac4a2ecfb2dbbf6009fc33b71ee6e86f0c3a5fa4912","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/final.safetensors","sha256":"2c6c4d8ceb3f46706d960292a04dce8cc34100d39fd705e7d9c39cec910c6177","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/epoch-035.pt","sha256":"170dd6233ce361653789c322916b848e96816518b39c01516ffe9d032300bf61","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/environment.json","sha256":"96a08dcaf5c64947be5fd131684ac1dc8750e633bfb4b48a72691d32fbc8ed33","bytes":5168}},"requestSha256":"5f42612832eb7aa90b93ea161339afa95427bf1744b47b07c34c78ba1edf45e0","metricsSha256":"5ef2d61d599f17e6429adab11595a3faf607864daf2dad198d27df83bedac567","continuationSha256":null,"continuation":null},"sync-wsd-h20--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2452471","recipe_identity":"07922fc5b47e3f7e2964e4c181fccec5ce2e575fabba8a8497e1657dee05e8bb","loss":3.5747387670841677,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"a81751afd88a4f31556607793730d3560430ec8876b6eb3929e313a8a5800a17","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"942e666fe401d938b9fffae777b11223f808334af8267ed3e1a39b9fb663f15f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"f4f9aa99b61b205c2af140687d87fb0223e60fe302f5f934f38e669ad6026814","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"aa1bd5df55112d11df3e7176b71c0c0e31f4deca369c45fccaea7d37db6691dd","bytes":244950059},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment.json","sha256":"2573c80f4574b64ff89be6fe18f6e564d642438500a811ea1f5e75ec0596d777","bytes":5201}},"requestSha256":"9f287c263baa7d100674f94db1755474051bdf4469aa00d0a18260a34616f79e","metricsSha256":"139e77807406c245e5b9a76e813a6bed3fcac768171539f9fd1bb7e4d9da9e37","continuationSha256":null,"continuation":null},"sync-wsd-h40--0.004--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2456190","recipe_identity":"302c51f95f86bf4f3afae463701ed00417f7ed6082456399f91ddd3d50266493","loss":3.4755529848956312,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"55248489e65911a4de5611ac0ec47755dd2870e9746fbad388bff3ad4ff7032d","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"f0f66afeca3673ec542a0df931af80449823017f507d400623567f6af8db6547","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"b190f74ba120216b61eb871434989a9dadbf4c49396331f1898b122332a97ab6","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"f6297ca4bee09ccecac8a8268250bd6e671f178b9d779df78c1575c604f4bc4c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"41872da3f33c71ccf674da6c13c15f71a51b6190d53ebeb68576ab5c41d9fcd9","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"dc6b1b57f506ab299e9ff84adb93ef55bbe93ca99b7b418826560dc9834039f5","bytes":5167}},"requestSha256":"b80c04b5937cdf84da832f9348941c3a471c6f3b70a0bb62532c48d54d7500be","metricsSha256":"0cab8ec72ca153ba809e4141c429485739f9083a007619127e255a9bed20f256","continuationSha256":"6121396a5cb25f7b771b3c1c8eca661fcf583fc2f4c14147e5cc1211d9457581","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"41872da3f33c71ccf674da6c13c15f71a51b6190d53ebeb68576ab5c41d9fcd9","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"572a4b6662da09a1acaed066c568092a09bd75c78821eb5832bb5b6f7c1e7c19","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"54d6dc8e326ba85d05f0aaae6980ce6c64c251fe6a417dce217ef8375564edd9","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"302c51f95f86bf4f3afae463701ed00417f7ed6082456399f91ddd3d50266493","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2456193","recipe_identity":"d639aaadf01a6c3fdf9501925d52e937a7f30f2de708c7404ece5a0c6340bf72","loss":3.4788332237586945,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"020b0562620ed5841148d68d9af9d0e7ac05e091eb9d963aeecd6e73d7fe066f","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"5728ca946dd0cf5a173c2fe423f3c712b36bdf398bcda4269b6696513844f20c","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"992295cbe1b832691325f06ef17afc5865b05f2789f963a9487faf62faf28aaa","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"d20d849f8e4225d9b9ceb150e0f0052bd549bb6b2c9e3a0ade8f16a1c531396c","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"4ff6a74a03cfa574da179023fb68767c2d83adb899c401feb4a1cf417746949c","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"cb92b8beea78996ea51e4e26c29a02c1983941977a8196b7fb24d334898a1008","bytes":5196}},"requestSha256":"57b6c4b7a3a49a553394719dd02fd5bef3ed7b7ccc7b49e474c2d82dd944e74c","metricsSha256":"ad81e6c6b15d239a4386acf7162db43eb225633a3d373e055f0593803ad2235b","continuationSha256":"fb1d4581150c2a2566be9baf886d987377d63bb5a39626c838ea08781f5ef669","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"4ff6a74a03cfa574da179023fb68767c2d83adb899c401feb4a1cf417746949c","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"202086d64a3f222417eb99afe6a2475a8b69a5b4a90de7fd73778059384e277c","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"535472da4b118a09df64401538c9c3bd0d6d6feca8c65148300a2a95ad61f525","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d639aaadf01a6c3fdf9501925d52e937a7f30f2de708c7404ece5a0c6340bf72","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2456200","recipe_identity":"f936f71bc4dc4210dc4d132b253b4a2ef3ae12986fb1d05de8133806e4d9e5eb","loss":3.4777120132638637,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"84e7882c438db4ddac0bda384a5c1712e6b182a6ec5295b70f30bea27dce5741","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"356f6278534ce084f947bd0f47de79dd94edf35385717c0e9ddba1093136738e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"4f82df959951e582852277f59d53b5358ff2d8bc63ac03b61c145fda7a958839","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"cb496f2e1a09f6a8bdfa8ba02388ec9f1266688883fa025b0a791464dd17acb7","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"01ae22a7bad332ef6b527fd1582f3b9c221fbe168e48bf7800db4108133063df","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"8d0aaf8b953f2725a2754b2191cac6d59b12b5213360d7f9121b301936a271e7","bytes":5196}},"requestSha256":"27b9e2ec8bf5f6023c905ad659468259b6587f9f86ecbcf32eb9358f88d8007b","metricsSha256":"22bf1b690ab61feeb598793a3b8aed7bb2e25d448f0b2df9421ed8ad2741a2d8","continuationSha256":"2db8970fe0d5a632a930f26481be2650657b1e5872ea8bb056ef04d66ad7d932","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"01ae22a7bad332ef6b527fd1582f3b9c221fbe168e48bf7800db4108133063df","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"da833a48046bf210b9a5a8f963162e904a75dd1bcc8303dc086912575519c694","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5aa9ba7ee236df30fa8bec4ddd8c4262025c68c21fe5e626ebad2be6aad5fef7","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f936f71bc4dc4210dc4d132b253b4a2ef3ae12986fb1d05de8133806e4d9e5eb","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2456205","recipe_identity":"4b12e59aa6a342f251b89ef63ead8a8e1ba5e92cf88d70b6db8346d68137769c","loss":3.4843115747706386,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"8ba1e74a236b01d65de53c0c85029b0e9bf08915d8fe161818225c2910f9048f","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"c23e65e30266ef27d55609fc2d695f502edb2819bc91a0b3ac494deabd75668a","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"674cb2ecd82a2a3f98b8fd80e62e24566ca09c064fe46209b1d00ad39fe03e5a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"f8786490c314af7f3b9ee2a3985a2c7751012fdcb48cdcb93f6336d4303bb33e","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"facdc80d956584f8442712a43e7ec61623544b80402bb343f02b3b1a195a3543","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"4a3b805265d32754aa0fa4179daae28a6a3049153676988b2ed56fb08a51edfe","bytes":5168}},"requestSha256":"b6af80a51b1d0867de47703f9f1b840cacb98cb3ae740a029fcb69b5fbcfe8d6","metricsSha256":"f1590d2691230796a1be0c373413023db0bf1e309e0a93d5e64edde1dc8c4e8f","continuationSha256":"03ad4fb0f9e0d0abfdc880ef49f71ecfbc001966f0b3b3b5927ebadc8895e27c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"facdc80d956584f8442712a43e7ec61623544b80402bb343f02b3b1a195a3543","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"6dfa797b1069bb89d57988ccd87d14021b73aa7e2bcd046d2a901ed3addc99a7","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"800d011d12404717311dbeb5c3a4558cbf0faac0f40e68a3f656ddb1856bc9e5","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"4b12e59aa6a342f251b89ef63ead8a8e1ba5e92cf88d70b6db8346d68137769c","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2456206","recipe_identity":"c8132a16f74661dd98af3f377f51f5a9bb4d1a7f343d95ded98a5d8c515505f8","loss":3.476859046660619,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"cbdccf6f2ee3c4d89f3510bb063db9234e4f47ce1fd8bee6d3fde8e093261d14","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"7e76649d460aa236ac21dd1b8b4b530f6e5ab6c08b8058f8bec18cd43b1eadb5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"88d658b6a57add663a380e8c865c5d2d400dc66350993260e69abc804bb3642d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"516e6d80190f739934bda93cad5ebfd702243acfc00f77c7a371d8fb592676a1","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"db583822e406beb943de02932deba7ae10fcdc084e7d0c2dcdba6664d3a62847","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"7cbd281966b2d5d3a0183209f31d5f289e89573cd159e94eff3124d347d168ed","bytes":5195}},"requestSha256":"5101f0c69012d6d2fe73cdb22318b61d5748c33380926ce8d5ca9fe07676c2e4","metricsSha256":"0545ef7353fc693773afcc46cb2fde7cd5ec7209eef3fc3670f27e16d6bdefc8","continuationSha256":"7b645680eb23bfcf6648e9967b297de321944011d95bc75de4a668523a8d4076","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"db583822e406beb943de02932deba7ae10fcdc084e7d0c2dcdba6664d3a62847","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"10eb84cda0afd0fad3000ec6c049bd8e0da8b82e0bb83daecc85f5d6687adb56","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ec856bd6402e2c28c29059d28d70c3c70786159194908429a57ea1dc199a4f39","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c8132a16f74661dd98af3f377f51f5a9bb4d1a7f343d95ded98a5d8c515505f8","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2456207","recipe_identity":"8b583b6c0fe88384b694511e6bef89c5ce1fdc7815e8e22fe4386a6233792596","loss":3.4856309002882964,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"89ca31289f56925a761bab9e8ce12c3c970ba594352a5e6fda170a7b0b6f3247","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"e3e2689fbafbc197e73c15b402880aea108a09e25867227130559c4f62dded36","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"3cfab763ba7463946476a3fa2ac72a9a8f89cc538a8bf64778010c2a8d81dae7","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"5e5a879f1d606e95afb158837b3952e296c08bda8503f78ebc6415c317ce7e8e","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"8a4f2db1fec408b0efe3d9316becd22e35df37ceabfcc840ee30599a198af0c3","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"a8a01dc42f0629e803934d2d2b3fd78591b679676d1d11a4528f6ffad8e24d3d","bytes":5168}},"requestSha256":"400aae476c24fe8e51d4965bc268d9eda9370dfd8018583693d69de2a046c4f3","metricsSha256":"1266a11de0353417cb57efb7a3f2a4831882ad2b6be9ce7449cca3ea120e65d4","continuationSha256":"b4485bc802683c5921e14756814f014b9940c39e593a947d9e255d580c490b58","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"8a4f2db1fec408b0efe3d9316becd22e35df37ceabfcc840ee30599a198af0c3","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"ab46e59778a419dfbbf66217bd3ce2b9002bca5b2a7019721ded41b10b2de61c","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3b48d7bcf540e9099ad1d9331636ad6d2f1ff870131ed75d9a45ad9ddc7ce6cb","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8b583b6c0fe88384b694511e6bef89c5ce1fdc7815e8e22fe4386a6233792596","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2456208","recipe_identity":"60b6f06523e461bddefbf4847990b2724cc01024a051634648a79d558a74dbd3","loss":3.496402676426997,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"5dc4cba3ad54f85be51296fad47602e3d5046258fd56841934620099eb9fcc18","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"f77861404eb58add91795e4d80b8c4b1a2e06bc80f26ecefc506f7134778ce71","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"8524d23c27a5c54a72ecd030dcbe46ff770b53afdd1a6de316b80b5081e25118","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"bd01f17232a927c208cea0ad427a058be11eadeb637506383e9d43f81c0dc0d0","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"c1737515403c4ad9c82b58eb273f761e6703367927e2df716856e364c8e466e4","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"796ae4eb4926bd65815f868ce10785bc69e59f3e897ff91a1adfb6d2eaa82531","bytes":5196}},"requestSha256":"da2be5bcd9f27189df0d5cdfdeb77a7ed3b4060e1a90719c178f33001d7a5a8c","metricsSha256":"46bc53521ebd44b9450beb918f15843a79e79e9d4fadfa15ba972adf1ffc7c9d","continuationSha256":"98b5d7f0ac9be66e6698f7cac0244e8cc953a38daf44a6703e3ad8d5cf6cd9bc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"c1737515403c4ad9c82b58eb273f761e6703367927e2df716856e364c8e466e4","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"919f2a5dc66cfea81146468ea4cdb115b568d97b7790bf0b450e202c27b4069e","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9b5b93607e0d54dcd5fa03be06d5e503477b4d69e007b387d9286519a324d77a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"60b6f06523e461bddefbf4847990b2724cc01024a051634648a79d558a74dbd3","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2456209","recipe_identity":"244052dab9201ece5211209c2bbe18700cae4838a91b091e4a1b313c2d51f79b","loss":3.484004256418404,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"18142a336eb0803b3abedf3c01fcf9f3c3fc5581e7f2a601abb83056ffeea7e2","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"54b6cdbeaa858fd5341436ea955512de4e58fabc1e9cd29f83844ceb3e2493bc","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"f3c85c20912c96671ce6f557fa24619ac43b04711cb3fcfe68b41198e779185c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"2811bde8272d63e4a010b68f8d14d3f39481aca4a4b71e4f47de77bf7989e421","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"b3abd46c3b78afa17fe7f9d2ae949e1ea2ce64e7373f33ff6e7fadb4adf236ec","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"74c808c0140f05047e80c316029aea05bf4def889b05e36909b46c28ed128661","bytes":5114}},"requestSha256":"022b64539afc60afa72362af0573c8282b5e4fb3c4de3b0c83cdb11027d40446","metricsSha256":"3125ed05f1574fec4bc73c69b7114d6a74a03f43051bd02c0b7b6326e8f3fcc9","continuationSha256":"59b2ad3ed5474867e08860f67dc31ebfa59959caf362d77abc94b00255a30c4d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"b3abd46c3b78afa17fe7f9d2ae949e1ea2ce64e7373f33ff6e7fadb4adf236ec","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"00956644119a6d56d65e1bda39b42209b07592a88424fba5d7821883715e6a94","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fb58acd1e3e8a31c52ae65d7659745e904a95f38da1c39b562959add19f17e9b","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"244052dab9201ece5211209c2bbe18700cae4838a91b091e4a1b313c2d51f79b","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2456210","recipe_identity":"1c052013a580b35d21b6f4c74b6206d9b04e43be7c80f95448476aca1eb5ff0a","loss":3.4881544762436234,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"62118417655a0e68b2b2078fff403043149ddd12a3f108d23b617f63f6d87bca","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"daf1f80ee2c9449107859ac762af30c0f5e8450bf6c84ff5579995958f1233f3","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"3b7b0f0a2b5d0dc57b669a2bfa3b66bc2068e0ce9aeab03e387090fb5baa5af4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"7c600467a29b628fb160fb781fe15c64a01bac7742fe964e9b92a95a89e3aef8","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"0d020808fb733c39655a2fd556644ed9620f7e83213794be7f6731100c5cbd94","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"6a3e1331908b99b941899501bb8ff51312a69606032b6178f9cda3364ae2836a","bytes":5116}},"requestSha256":"95285851428c245f46a96da0c6caeaeaaf9a0c46feb5d0a51b26eef7913135b1","metricsSha256":"4d9882c88593bd426336bb69a2736789bb97d6237e30dc511b7d0aaa7938939e","continuationSha256":"6a4a861a0cc617ff136d29a460499377caeee02a6a450dd8ad0b3309db262951","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"0d020808fb733c39655a2fd556644ed9620f7e83213794be7f6731100c5cbd94","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"76142ee313da41b3d0e459b8e8aff9873cec09ee6f9fe01d95978d8c0997c8b7","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"7e33810613fa5bcbf16e5236d958221dc951db67b3b27e536a9ba77ef1c41851","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1c052013a580b35d21b6f4c74b6206d9b04e43be7c80f95448476aca1eb5ff0a","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2456211","recipe_identity":"2f6405e8f779bfad9dc3b2579c9bc7231f7546e066268c8cfe50e51c41a41344","loss":3.4548085307638963,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"7a53400711b8799bdb353da3a68e446095d7fc594decccf4631c6b55e7b92460","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"c717d65b77a06e09be5b89a27f8c8cbe57a89ec2485b8c56043cb31f18dc5384","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"c12afa0ac86dba87816c6400b8682747a9515e41b849d1ed270724f32a60b3b0","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"974a85727053363e417abe173b0b6d52f0aa6334f6592afe976dba189b4a519c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/epoch-035.pt","sha256":"7ef4456c4f99b563768690d3058be4c5fc5d2ff4e8f7d44e1a7a32523458557f","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"f574e24f9dfdb5bbf66af69edee6b0d5e054cce657ad185cf4ced20768e194b7","bytes":5173}},"requestSha256":"592f98d6b71fd857f7f803a65357b8043046b8d6e46d5db74452c7a58c579a68","metricsSha256":"2b48b60a1133a6c45ec98ab308011b1cba49a7ac263014596491f70dd180ff8a","continuationSha256":"026c0eaf64b1750bb3adf722fea7cc54348d2e72143cd26cc3ca39f34c574afd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003/epoch-035.pt","sha256":"7ef4456c4f99b563768690d3058be4c5fc5d2ff4e8f7d44e1a7a32523458557f","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"0b40a8b8a1b3cca72ca4c5c1e931661428e319b29475f0c27605c6ef71826479","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-003","cpu_threads":8},"decentralized":null},"recipe_identity":"fbfac5d6f7fde07f4d3a17d75cec051ff6bff880a43c9bad8b2b816e955171b4","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"2f6405e8f779bfad9dc3b2579c9bc7231f7546e066268c8cfe50e51c41a41344","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2456213","recipe_identity":"8b8e9fcc0dcf358804bcfc043a21c1f43474cdb9c368077503826d23008ad177","loss":3.4679864698303144,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"e80904d261c93a913d19d1bad5db93a15a28fcabf3f0f7e10be857c6c6b497ef","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"e28104806dcab28061f52cbb8d004c8c8f66d8181dd641053ff27083a07a6fd6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"f483ff4eb02c30c073c7ef4e8a9c9b94f6fbd9f49f35b81441e33ae849d0a2b3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"a835f56bc850ea7b553c5469a918a5194950260dbc073c8fd02505fd34708cd8","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"c2a9cccc0002d653609bc66416f2051c85f03161603d3c5e6ea49a72e6b5cb60","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"5a5504ecf324d026444c9a9cb42e14c4ea5888495de51df0657a04e11ecb4655","bytes":5170}},"requestSha256":"1147febe1746252fe4b3332ea1df910a640fc897a9c8acec2e2818a273e6a457","metricsSha256":"d1b5ea4ebd4b3a44f364d9a1c456f9f0163943bbdfbc2d728fb6fc4da699827c","continuationSha256":"208923642f13aaf0af3ee6f9f04681ab0d3877c1488d2a891fc9070f4c590d0a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"c2a9cccc0002d653609bc66416f2051c85f03161603d3c5e6ea49a72e6b5cb60","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"c55863b09ad79bb83bab3384743a10d735be91fd297e3e28b3bf8203a3ca6704","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"244a0a4f330887756c78df3e67dd685586c23e8576d01dfcf90e908a03d10f1c","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8b8e9fcc0dcf358804bcfc043a21c1f43474cdb9c368077503826d23008ad177","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2456214","recipe_identity":"97c6639848cb4eb4b38489a677a87b5535bfdcf6f4ebd56f824f3d6d39efeed3","loss":3.4673522786949325,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"a428779ff2c96cde8f40c7bf8880883d3735ecd1fc678143f9abce7d8b88fe59","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"9c13c7746049da991b279a198420825f437c88870832c5b993225327b2ae6273","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"78120e7093f2bd8a39d56f570fc961f2a90653326ec28ad70bb7c105c263d541","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"2854291f5978b27456d525a2ed81659ee8c3119141eae0183db0a824516d9e1a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"2ad79036697402fe621c1d8e0ec45d1a2d66ed27af15be4c088f590934251480","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"078b48c4eb93721772df0a45963af0c4cb180609caa3ad6f13aa854a774cc2a6","bytes":5199}},"requestSha256":"6c28539dcf99beca6afec47bc04d22439a99859ecd228a4b10189deb9230b0bd","metricsSha256":"bc83bbbe6f9257357894896ed771efa0bbddec8e5f2b4a151eebacc25d2608c7","continuationSha256":"0f9c0c9428c0c181f2fbff11182660fee1148051c7f386bd905bee26699ab68d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"2ad79036697402fe621c1d8e0ec45d1a2d66ed27af15be4c088f590934251480","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"ad1d9f357385f34a93d6b0f3a3442a2161f7d11bd7fa64f341b61e7181b4f8fb","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"460b5b352c7281894b7d5e954b964259a6982885dc908b1291237dd3f3a33ccd","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"97c6639848cb4eb4b38489a677a87b5535bfdcf6f4ebd56f824f3d6d39efeed3","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2456215","recipe_identity":"c7b387838d2b2d324bf3efb81059fde2d66709c61b4e7b495a211eb58866f6a6","loss":3.473158218724192,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"996a8d7691054cf08b97f764dcd21013304ec99b83a249ea22218cfe10d271cc","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"86b29a93585fd87e40ea0b04030d78be2add39ed85936233bb82f44e4c2a5f63","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"600e0cf8709c0e462d6926b70e41a1c42a3a11eda3f25db2665de3e671612691","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"2cbbf575fccc7af0dfe8dca370f92f80584863eed40b9be3420dd7a82710737f","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"2aac8d83b1c08a3863bfcae4200ee1c569b06006c75eabb33ca98fec25b84c7f","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"9c96a94f0fd64cb525c333e7ac66164ed8fd36e797764c6d348ed68e54414083","bytes":5115}},"requestSha256":"d52edf3028528a3a8857dcdcd280f5440ca9956c1893f69cfb75153fb7fc2845","metricsSha256":"2e87731985de7c6c92f0ad821d1a07a8fcead5d6bea596e3a06d3b390f28f83f","continuationSha256":"dafc7aa294bce3c8d1a3e07f322c87f4fe07790169295d3fdd0ab68a9c6371f4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"2aac8d83b1c08a3863bfcae4200ee1c569b06006c75eabb33ca98fec25b84c7f","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"d8bd6ba1de6f01a63e5df3201ae89dc9fe46184675a7d32bcc2e85fecdc6a285","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"dbec4bad17a92472c0d1f38ce46ee688df78d35aeb09c20233b9b3365b2163d8","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c7b387838d2b2d324bf3efb81059fde2d66709c61b4e7b495a211eb58866f6a6","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2456216","recipe_identity":"8d5ebf9ff14db86a803c1b7294672584ea1b8d41cdd62811b5125b477f1dbc4d","loss":3.4681682661046906,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"43b0a2b1cc83ca4183f460ba3a8462b65af6bc6b269394b77eaafe573baf7dec","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"b553f325a6911e965c1e275c8137f79635cfd79e7533d4425c3f6593a2faa498","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"992f6a126c3f3a761c0d0be444ed6f297237c1dc8a94afeb9d6a48349c5c35de","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"80ff3af5369e59d8be8e150ecc00e75a234f1c2fdedf4dd31cff6a90c7ea24fe","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"d659037d3dd282abb4a0e8879ede0cfa61ef738ad42d550f51103cf8287996f8","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"234d4ae73361cf26d5d9673474c4335722962dfaf110de1280f1667b06ab4df7","bytes":5173}},"requestSha256":"e99b24cfae35370ef3e9ecc579f90b24ebc86f473dd6d0ab74eee9a4e8ccef2d","metricsSha256":"3751ddc8460dfc97855ccbba70fd250533aa8e2d4591f60b814de6699ccdbc51","continuationSha256":"424364b4fdc71516065988cadeb72ebbf7164d7972fc7539e0831a56138b4d23","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"d659037d3dd282abb4a0e8879ede0cfa61ef738ad42d550f51103cf8287996f8","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"b14cb6fb9a869000483da6818c70b587f722d0b496c40bfdac3bfccc440f30be","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e9afde9cbafe9533ae8a758d741bac74fc0a95201f617946ec567ebd5b076a11","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8d5ebf9ff14db86a803c1b7294672584ea1b8d41cdd62811b5125b477f1dbc4d","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2456217","recipe_identity":"3888794ae6a812a624d32d79cbe9b18a96408624bc711130e4ae7e4a7ec13370","loss":3.475472607358183,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"6e9db37f5acc6e11224b04246b0ff09d1a131f4c20ce4c0a9c3f978e1dedaed2","bytes":738},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"2f8df57f2b83a3e898a4fcc8f4b8470545e3bd1c524d67540046780202992c08","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"36b17c2cb17566c2142c81ddc29a277de55118500114f35054097fa0f921bd9b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"5436f9df25670c61dfd9116bf460871d51c837f531b57ac85b2a04a63761c15f","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"0fa944632d7ee2a955d762d5139ff673aac2cd1877358bc684027297aaaa44a3","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"d0f8e0512eef75c512e68022d3b7d96a48675f3b89a3ae34c86a87cab010925a","bytes":5173}},"requestSha256":"8b1fab439f89bf5a8db235b9ce45413a37ca85940b079a6604a319febdff2e30","metricsSha256":"5e96f4f68ac4624293eca4a87ea16849d87a1d07fc3d653c7e554884ed1f3365","continuationSha256":"29c5ae8259263a2a7c3519abdc07cfbb96e730909211763d1de395fa24297c98","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"0fa944632d7ee2a955d762d5139ff673aac2cd1877358bc684027297aaaa44a3","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"8b74cfa10bdbe3ee8aa45ec0abd1fad5a208b1a82244e22389823d62619d3d64","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f64cefe8c9e4d238872951fa098619e49fd72c6e37f2a12ebf119beffd5031af","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3888794ae6a812a624d32d79cbe9b18a96408624bc711130e4ae7e4a7ec13370","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2456218","recipe_identity":"418c4ef9f43f2bde0daa6ee28eee9cb6da3a015a126564844aea51931abf203b","loss":3.479317315960111,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"12954f35efede29b439bf5de55fa6fb75c0c2dc4c275d0519cb7d2882f34b7a7","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"436ccf8d069f8bdb7c9a2008a0fb5903a510ea063413a9f5fe9835ee412db5db","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"16585d195e124044c54b0daeacb91c84b1c3c604ed33b49f1b4251c7a9e0efed","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"f12a4f8dfb57d40ebb207412090a955fbdd2d03bc5e63e7425d0f5ec9f4748d6","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"054982a61b8d9ad9154217d90cc54bf83db2513df72430a3f7bd68bd56554b70","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"647882a547a84dd4de058088833423be574afcf90e1bed79d129b6cb57c7ad64","bytes":5200}},"requestSha256":"229db8f9506b0993ec8f1b32bcf23c044f1a95d35f717e42bc5f8d22fb5a32ce","metricsSha256":"be88d03e3a31af018e0bc5e12da3efbcee43fcdf99206e5f03e9efb53d01b660","continuationSha256":"cf720cbcaf84253175702297a01cbd9f87b20fc25e1e5c048eca655972ca97cc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"054982a61b8d9ad9154217d90cc54bf83db2513df72430a3f7bd68bd56554b70","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"6d90177fda5cf65d059ce399cb14a5410a923ea6cee5c10b81010be8315fec22","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"bfe6f702a4e125e905088e5864c21d042176fde600002e5632562c7904149cae","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"418c4ef9f43f2bde0daa6ee28eee9cb6da3a015a126564844aea51931abf203b","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2456219","recipe_identity":"2c1360c99b9dc7b32dcbe5a4e8ad4b703ff086f008fb45a44ce9425071c2d45f","loss":3.4667723287160865,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"52c28a2d329d397a5b6ea2e749d25af18104e2c37059e672564d0acdfe656bbd","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"8441201c8441a2dec4ce63fd43f69722f54e6a6339e22d9ea8098ee576746528","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"9d6ee78d9677819673dd2b86f08ea51f329e60eaafb853a82fd0262e1fc5752a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"a2ec180b6ef552fa2696130a04cadec6d836df2308042fbc9bd7e9feebe4f677","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"e754c46bc32d0b8f5f9a56bd680a627c8a0f6b0f89a789c4012f8017ba6b4b06","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"95fe00fc826d7858e0809230548f2c6521b307e2819a3172e6489825d7b208bc","bytes":5116}},"requestSha256":"f04fdaa81508e91653123be0e29886dfa151072412393930714be1b2e275ce5f","metricsSha256":"03b2b4ad1c7730795a69165738dafa99fb001554d5911c0f70aee88a1b32b5b8","continuationSha256":"3f55028835a2ed258fa95eca6d5b1ac76881148fd4f3161d7a163ce637e08ada","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"e754c46bc32d0b8f5f9a56bd680a627c8a0f6b0f89a789c4012f8017ba6b4b06","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"706e18bf583b012d68574ff370cf5193fd048853e1d6e9943c39ab564cd29441","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"59fc08d263067e76cfda5d21298061c03a966803c1a315d06147b737d2a6ccc7","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"2c1360c99b9dc7b32dcbe5a4e8ad4b703ff086f008fb45a44ce9425071c2d45f","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2456220","recipe_identity":"6c46b6f57ecc24b4c5ddf4565637d0891e5948119b0b330576f011545ccf9da3","loss":3.478118580700111,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"825461b7dc7b087eeaef6fd48854f04bb037357b96f1a54a6d6487c2a5f04516","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"816306d1434f3ba3bf7072bcf33ae5a37d039e2e0d14886993f0e61fbcf98a52","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"310325a3ded3954a922bc2e8fd366ea1a49b366d89e0b640526d4a30133fa195","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"08f12ed051bd0c37e9563630f60bd3a625099fd4b14d51dd7a5e53cf49f4502c","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"85612c22f2a172adb68eeda4581dbc5f6e4e91f17360770bb8b9d82d5172e7e8","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"17525cd466ff3dfb580250fbafbb27cc87412ce5f812614aa338bd2d77488c17","bytes":5173}},"requestSha256":"90ac27e72c77c64a8bb4a5953b7ff52d5f7f1f38a915c6d5df8959764cbd86d3","metricsSha256":"0d9d8430d9b453666706a0531540b62f77628de77242ac4be55e39f702c0e3cc","continuationSha256":"bdda042910fa86829f82580350741ab9a88f4e23485b9ebd430fd0114b48c6d8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"85612c22f2a172adb68eeda4581dbc5f6e4e91f17360770bb8b9d82d5172e7e8","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"6467a36553a2da220715a33040142dba373a7ccc9fc96cfcfcf0c451d72d71c8","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"30d49c03e5e0ecc9c431f49c252bbf98896d26e41f136d22d61b391045880886","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6c46b6f57ecc24b4c5ddf4565637d0891e5948119b0b330576f011545ccf9da3","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2456221","recipe_identity":"6b3197b0c883a8e9a85ecf7bf586da7fe161c9b1438367331756abf7789e9f99","loss":3.4735047162075072,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"f21fc6985aab3e5fa37ce7d07cab64c81738df38f5efa5bb1b99ae1ddec6d070","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"c2ca0c84bb9ef2d7e4ad876118e1625e976eab7b1673736c0ef84108024e8264","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"87a734d96c1c4b0172541cb427f1b4a9d1e073be4bd3255df9f4a650f6eaf7bd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"c63f7fe04f3de7dd1edcd131bbc7d673010026410f38efc75c69ff85356c6c47","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"28969ba28acc8d42f856203eb6028bed4552cf1f6954d561870af43547a796a5","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"7e3b97ab94ea27fc092b0ba69d8703390bc90f62d0c0f43475e983fc0e4c9dfa","bytes":5202}},"requestSha256":"c18f730fe17e3f9c4412006dcac51d615af050bb7fc50ba2d4a9619986fa4095","metricsSha256":"06970c445557948cd5b6a32d254686a1a14ddac28d9350488cbe80229fe570cb","continuationSha256":"fe9c1b50471e4dfcc74344ba837927c68cdbd1354c4615b2e658488839971355","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"28969ba28acc8d42f856203eb6028bed4552cf1f6954d561870af43547a796a5","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"fbd9e7ba574efff3f866e91e3923f6439c05add066e82951a4a2f56353af8a94","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9b28a46ecb73b427269826507c2f3e3f167db0f5ce2b5bfa5d58cd857f4732e0","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6b3197b0c883a8e9a85ecf7bf586da7fe161c9b1438367331756abf7789e9f99","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2456269","recipe_identity":"0513547ebe96c2504ecf688bebccbcbae61fb1453e075f6d64b185ed0742e177","loss":3.492706443056062,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"f7c92b757f9d9da5fddfe33b337fcd8e1047f5c2e2227dff297daa08ba381903","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"c6e83b9d8437f840ee9ddbf9a9f758c703ae070af514507c3482cc3b6401bd92","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"40a122fd6c7e8e7018b02c4213707a9f260fa8a2d4cfdca0e3d427136f6e61fc","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"cd5a3c3badc3c4c854f535422f1ad5f7fc3d984ccb98e5a9374c88910bd99e5b","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"26e4a43dc2d6c902852650c8a81f1b3de5e14065c68f9bbbcaa156127fe7e0f9","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"46a12e1da208e60e844b0eccf9276ff4276ac24040e78b8b132d6f0ed86a8a0c","bytes":5202}},"requestSha256":"2984ab6d764c8dff0c821f45c8ed515a0d69f5be7beab04715d87993efe6e9d3","metricsSha256":"69977ff968199389a2ed5304a2bc7cfbd0175bb218cc1e04811d8bbbf5959b9d","continuationSha256":"0a28b184f0d5ca89d9b4e9c94c56d4059a68ddcf4816a4b8127e13c246d5b6bb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"26e4a43dc2d6c902852650c8a81f1b3de5e14065c68f9bbbcaa156127fe7e0f9","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"52baf472d4c3356717876c45c3bcfc495fc7534ee181ca8e9d291c60c9bb65df","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1a0820bc550260835a8d2ff245ed6f7262ccc130f6a3293a712783e94501dffd","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0513547ebe96c2504ecf688bebccbcbae61fb1453e075f6d64b185ed0742e177","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2456270","recipe_identity":"defd4aa99cf6d2ac9c12da1efe0f74b5d85898ed15457b9c99c7aefc69273870","loss":3.470226172025489,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"61adbbbf6d49a3026c38039aa2cb64e3ac88aea3693ca28cb8e42e5b3c9a5b36","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"e9d4ea5f66063d618e3aefaf11305cab7d2a6a375862ad7e0a4ea35c37755f51","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"7d9700158f0598002f2bc2d99e8defc49231d46061a9f6635bfc01b34a4db1ff","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"e90cf127e310efd496c17a37662146722277fa250c254a8af7efddae702c1965","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"59c13acb96c4f6e68a8ad44dd904531a79c7d05c5e4dd585244b6b9d848ddedc","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"a09ddaa8b770f98572425d9c18a966cf1a034a3c7c6f3a085a7f8f5901bdcf43","bytes":5202}},"requestSha256":"f2673967b7ad3e6f3df873d7b3b8bff6415051b50c93ac747394402630f43fa4","metricsSha256":"49a25d3c44dca5e05cb5b6f7fbe574fe58f944af79eee9ea72051c8a1cd9fa4c","continuationSha256":"1f2c3ddf4e21b3ffcd068fc4211b48f63129c1181e11f9c2d5f0eb20aa39f93a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"59c13acb96c4f6e68a8ad44dd904531a79c7d05c5e4dd585244b6b9d848ddedc","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"2198c2d97f75f18b08d893fb634473f6f20701153fcbd789d553ed8315dea27a","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9ed724fd6d66dc1d270cd7502700ffa9b3c6ce23592cbbd592f1cd0f492995fe","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"defd4aa99cf6d2ac9c12da1efe0f74b5d85898ed15457b9c99c7aefc69273870","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2456271","recipe_identity":"d2c8414f2fb82e9a7d1cfc92746d4582a5e8776ba107a6a7a5757db70c9fcf79","loss":3.4631896312463604,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"b2e0422e561c3979b8c9a67e64c8e19a8c7f4d0092a2a1d581fba6c16286b622","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"905936bb53050ba46cfbcbe932dd42ef1b00e0c5ab6cdca62f1f90515177ca75","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"5549067fa303fe9c570b9c12d6498db61242e289f6f4de836db4748b7acb9a3f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"7b6510671b51dede416f642c61d7c7685e3877d0017fef4f615a4b31a5b5e4de","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"e978cfcda23620d19ce3f8330bc41252bf3b3ccb2b52001324078d1ffc9d319e","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"570d8d5479ab0d585cb1dd567a041e7a7ef4a14264f3c823fa79d0fbd7b99c66","bytes":5113}},"requestSha256":"a3bdd08ed4eb6412808d17477567fcd1b04f8e8b8e1cd110ddf0d918b10d3203","metricsSha256":"030ac6a1ff6c4af65d5f2fd20beb776dd989253335406bf239475b9df18b3840","continuationSha256":"16b420f17269db530110af22c50739a0cfc2326f9794babbdf5e98de8a823115","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"e978cfcda23620d19ce3f8330bc41252bf3b3ccb2b52001324078d1ffc9d319e","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"a7a4e24af9a6760b01b0dd26af2f2e6afbc3285b6e8e63c456cd58a8c1a2e878","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"da53fca47fea2be4fa03b06a5ab1d2a3c1be3fc854325cf0ef670499bc7dc593","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d2c8414f2fb82e9a7d1cfc92746d4582a5e8776ba107a6a7a5757db70c9fcf79","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2456272","recipe_identity":"f0e5394c389af0a7893f5826b180ff9a4ac9451394c4337712941a271008aa23","loss":3.476197257843589,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"0c5bd146c8cca3db57a79941320d634380533df942f9ca93944c21641f2c66b6","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"c9406277e2c96d0932b4025cce0b7aab51e73fe877e97e2ce318016f3f8d4c1e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"9733254c24af4ee57e0e40ce34a2fdd835541c5219dddbd4482ad53262319200","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"4d8b73aa47819a794c0b1e6ca30edf3068495f40c6a61d8822aeb9bc23b684c5","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"5d1e0e20001eb3b637269f7ce46d517baa448d76f91bb39a65e9ffa16a0c8cef","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"ac5d44c3463f8db7c4e560791a55cc05b5b79f76bd27231919643eafc289ddc6","bytes":5199}},"requestSha256":"adbc1922e74373037eea7a3e52c5e353876f3ba1a7bf4b98049055ac8c29ce08","metricsSha256":"62b30693b1e2e64b80f1005c1f1b8a29d2912982d25b3d17296e0d7fd4518e69","continuationSha256":"2f351f1054cd1fcc4fe5424e52f4f7715b587bf024cf92707dbe20993e7437f4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"5d1e0e20001eb3b637269f7ce46d517baa448d76f91bb39a65e9ffa16a0c8cef","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"73bb2b86d9cd9383c9daa62f22628bbfc1a0ee3f96e50f578efa1a929da18e7f","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"840287bd4ea4758f204c763cae7c0075e49ca3944c801fa18c6f60d883d4104b","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f0e5394c389af0a7893f5826b180ff9a4ac9451394c4337712941a271008aa23","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2456273","recipe_identity":"e1ae2184f4151d958ddc58d860b413f994203b60c09e8464f7ea7ae7f26372ff","loss":3.4755693040708757,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"6345f0379918b50363bc54f16a02da869e337c58997678665d57d865ef5a9562","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"79957c0550283f7cd198dddc21fe0e8a1145880bdd78bbcd99e2fbbbd52c515d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"83b136317855e1533acabeb33f3e8221a84a6178a6692526c5fcfa1209148653","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"36c5b0a7d06ae5cad8ccee0060018bfed4796eb32ed04a317a057e9f7c5fd01f","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"037d5b34fba722819ae56f1fee9e773e0efeb7de869f9fb32e91aeac83369bc1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"90b93b981039619ae319ef218e72d828da2428bec9645192f88c35e4625797e3","bytes":5199}},"requestSha256":"4864227e0f57f58a145fc6ae6922dd28e23fa6e4a1d8e892d4459fe5b1bd0943","metricsSha256":"b6c1f55e9aa2663a87eab81f9087c606013b38ad25e1087e38a5cdf58d17d18c","continuationSha256":"a1d5c35ef3ad29f00446c57ae0ebd6aa1cc38dd6aba4882495d5b7609e901516","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"037d5b34fba722819ae56f1fee9e773e0efeb7de869f9fb32e91aeac83369bc1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"35b0c6072d05708d76edb7d695731b7ef6aefb92300f9043ec8150e88d873942","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5aaaf035810427de61588c10be5a435509199a408ba6f031ca30b99aef4b335f","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e1ae2184f4151d958ddc58d860b413f994203b60c09e8464f7ea7ae7f26372ff","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2456274","recipe_identity":"0c57936e6f97fb101036898590254a47bd259dad44b11ad31d242cf199dcaffa","loss":3.4824460882651778,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"b6a7592ce15673fc15257f1118da47ff674a01b160f8dd59611e991ea1706250","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"9c7438c938f5292e230906f97a9b551290a8d2349a9e7f5458545895e45f7403","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"6e951e617e62c317cc0d16462d54d09a91beb8cc1bee279ad5109ccb1ed4a6fe","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"afbd78612056018d98798468bbf18ab649c1f59bc87cbd20a1bdd77065f80aa7","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"b3440ae8d3c966c3727d49c4889eb570eebb21cf69cc13168f0d70dd2cda006d","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"1e42fe02c3984e0e2e06b35bfd2f0d99da1dbd466f2feb59941ee04d98e6445d","bytes":5173}},"requestSha256":"b3a8292b1fbfec547d477f6c529e9b1309d9e22410758d4a46673bb288fff56d","metricsSha256":"e65115033ec7066a9bd6ba1246f3df82876337730ee27e942200cff27ef7794e","continuationSha256":"9d7c27baa0fc4f49b3de8c49d1ae17899fe01b7ead2d5b39c77e175e929a566d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"b3440ae8d3c966c3727d49c4889eb570eebb21cf69cc13168f0d70dd2cda006d","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"640ff892b2718a1c74a21fd502f6fbed73e16c4b8d2f0e474332d67de340825b","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5ef8ad14a4a8cb6eb89a7f625c7e3b71a5a0bb4e5c40a32ca0881234d8fd3709","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0c57936e6f97fb101036898590254a47bd259dad44b11ad31d242cf199dcaffa","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2456275","recipe_identity":"3adf90d7b05ed2015a29f06942899da4abbb5b2e574414a900e2deb0cd0c6b41","loss":3.4771503695215995,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"ae7ff63e760d19155597318bdd38b6da126ab05839950ea06f81453a8e18aa2a","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"21a17c3e0ac136a6622c00a7df5462d5dd569c0e7fdf38c934ac653adbf9ba7f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"5b8bdd45251c140db63d6f9fba5eee4c1008af44fb0a6577cd018b74616de52f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"df0a127af1092e5aa723f78b8c2693b1929e9ffe1525e9d082c44ee9996e2464","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"24a2999ef4f2009ccea13ae49b38810994a6008ba27b884018f73340ae0aadb0","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"66e4714e6c9122b7a9389f4be3d1e87ef97bde4a7bdb656e72a65f43f7b165b8","bytes":5202}},"requestSha256":"cc61884f8eebc77c74aa6f90ddcb532423e05bf327dee77998b6e7703354b7f4","metricsSha256":"c94db5266e7c54a308097fb6e997cebc93588c1cac600c3ef4d3aa6901ffb730","continuationSha256":"5eec1ea764d86a373c7c5af5b0540b8a07e7ae21305564bfb0ce45eca9cd90ad","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"24a2999ef4f2009ccea13ae49b38810994a6008ba27b884018f73340ae0aadb0","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"69d83ed03cbc6c9590dfdbd318dbbf6d97815a37a4c898a6f7aa5cb3f733de37","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0f8cfc9dc229ba0cea536e36ba09d8c5cba7bbb5a6bb2b340c1b571ec1316454","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3adf90d7b05ed2015a29f06942899da4abbb5b2e574414a900e2deb0cd0c6b41","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2456276","recipe_identity":"5b08ed312d173a3a1101c0daf4c42b52217dbf1251ffee0dab557d4810cc09f9","loss":3.4767636819031935,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"7353b8d7af209e5426a8e11755b1693ff5e273f8a02fabe0fb06fe3e58e2bd8c","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"a6fcfaa2929d3933ef725a7d948296cb4eddd6beb87ead21de9c11add1a459db","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"5ef15cbf92191b8ad6f3fe90153069a78cf3c8dbe9f90cf9af3a04c6057eb828","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"e5e001d5fcca90974519fe86d590261f9b2a5b446e952b85a448dbe747b0aee9","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"ab03b960348588593cc46b62b351f6cf43e82c686409bcc07f95aba2846921c6","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"71236aecbeff41bbc96d26c78353ec512b6a3532e42743d91671ca7cb2a1cca2","bytes":5172}},"requestSha256":"768b01ab246810148051ed4ff80fb0b1cdbb4082ca7f096a8d55d4437532e8bd","metricsSha256":"f6b771b63c4c08032b61b918dede91cbaacd734c66515b7064181aab8d738feb","continuationSha256":"630497525217a9d31f1385f0c0501cef1416244baeb942d3a8554432c90f1cd5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"ab03b960348588593cc46b62b351f6cf43e82c686409bcc07f95aba2846921c6","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"2696868d120118a2468730e2ff7742cbbecd843ed45278db207d83ddf21d8402","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f92dc85f08c6a7a690497a280d6a347088489e34cc7a454e352a22731b12005e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5b08ed312d173a3a1101c0daf4c42b52217dbf1251ffee0dab557d4810cc09f9","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2456277","recipe_identity":"adf17e69c1a1c29050d4096fe1baa27941863bfecc98bd511bd758f66f339c3f","loss":3.457830407098489,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"f78f646fed5ac4237fca54a4818fadd2bea2591ee3063f1b806fe771d37c61fd","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"48d85d38bb2a475fcbd4e7125c972171954147756c43aa726bb818db9746ca06","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"7ba131f3080ca8ef899e938788ac5de2ce33be9d088d375a7de28346e2fef5b4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"fbdeab103546901e3773b06b99a538bf355bf83f82e961f2bf53b1570bd81652","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"45d0a772331137dad76dab88ffa682da8235831549139ecdb6011d49b191138c","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"6c5e66251eeda9dcd1426ae0078ec9678f867e0799fc73d5bc0c1941fec1df9b","bytes":5201}},"requestSha256":"2d2991e28b3f89ec57bf94c899a97d92c9daa7f9e44b216a35d2ddf0ce927323","metricsSha256":"a833bcb06fa43a2a47997432031b58d97ed834c640c009d1b3cf76f96084bf64","continuationSha256":"56c887ad5f75a671b380ee96b4d84fb4267993e1f303bc3a1796009b1af3c152","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"45d0a772331137dad76dab88ffa682da8235831549139ecdb6011d49b191138c","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"d4edcd0c680256ff0b5518160c695db5387363b5298c943a69c09efe45e5d299","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f60d722685c2a4866dff4e16586845a0b55cb1a00e1ea8e3a4f5fe2826515360","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"adf17e69c1a1c29050d4096fe1baa27941863bfecc98bd511bd758f66f339c3f","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2456278","recipe_identity":"645937ef2b90ed2f75dfb2df714cfac0b9740dbdcc58d708f40d1a1aa03f26a7","loss":3.4625445275647917,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"e3bff909efcfea751f52a5ccb17f2a66a9a0b36fb3011bb29b32a6ba3a5aa223","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"1f8fb30b40b8443e4f3e00c2e48c27c18beaba0b9a1c18035a1e230a318976db","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"b298eb17288f3f78cd2bd4209d971134b58871440bc975061d5e34c64e84928b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"8cac7fffe4805e413ba16a92d9d336186b27da7881485e55b325cb8b2f29fa2a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"75e3163d875d1f906607628cb06ef60f409138bd626fd4835ac59a4aba6109cf","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"eb5feb157ad8f91114845ce6d05687bf395ace0613302120e8a90d5f7966c6c6","bytes":5112}},"requestSha256":"3f5635dc1411fdfd5ed8ad8a3e367f00d961ce26652d2117e91a762df961a41e","metricsSha256":"cc46207ddb9013b4073549ef105a145b19c4d43325c142941c6143ff96582af8","continuationSha256":"531222c8fd43e86ec5edfd8c0e58493e8ad57ef3129bc5e0f37e23295a24f279","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"75e3163d875d1f906607628cb06ef60f409138bd626fd4835ac59a4aba6109cf","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"940280afa066a8d4fe91e9eda584b7ee50664c5e7a5ff337bce0b43a9cc844bb","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8c1e550e2f53d514b1088f13b06292b79eb7b5384915a73b24eef0db2d2abb40","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"645937ef2b90ed2f75dfb2df714cfac0b9740dbdcc58d708f40d1a1aa03f26a7","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2456279","recipe_identity":"5ca8d1ca93ed3e3da00593aeff0086482f03c5516eeb05d6ee8640cf06b07b02","loss":3.464490102473478,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"b8e04dcf478a614a7b0faea3c04a8650b49e8446b28497ecf7c7aabcf62d2102","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"86751b8e40303b3d9d9ba481ce76e8b578e1fba0db5767ef042b8a51d033c742","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"b6b69cef997b66c2528a9ae129a6db66598a65b42e1f9e9a742e440dd115687f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"4636c936a97ccdbbaf830b9239a2ba1a04a3bd0b45d5d79bb51b774de62af6c1","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"a58221a4cbae212234f50a9c566b5235e1d3fda80993dd0668f6fee002bfbde4","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"82184866764d17766e70e11566b3ce9b38ebff79ddb0c2a8ce9ed94d8f5b72fd","bytes":5169}},"requestSha256":"1d676f313dc6a606fa0fe206173490b0419c24fd14d45ddabfe66b9022055c2e","metricsSha256":"e72f573b9f3905ee095cebc892a53ba978ff548ff331e164f11f2b1208f87fac","continuationSha256":"6a6c2de7d508c1c036dec98f12abb0c4b3c1533f1fa333524f37be078421f432","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"a58221a4cbae212234f50a9c566b5235e1d3fda80993dd0668f6fee002bfbde4","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"3c2dc2143393d3158eec1eb5b77d7cf183049fd717ccea57ec6879d0064a61a4","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"eeab317f65d1d74056a9937d793fb495a528032fad1ec3126d2a18e40a7059f2","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5ca8d1ca93ed3e3da00593aeff0086482f03c5516eeb05d6ee8640cf06b07b02","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2456280","recipe_identity":"1e3b203729b323ce163b556cc93b9bf71859766b6d473f576585d51757e52de0","loss":3.4577991555613536,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"ee090c5234eb427450e3e7d0187e09ec8c55192d18f5dcb5873c04f467ab719a","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"eb2c4b64350f4b388472210abbcf550bb1fdb763d2138319a5067c2f7a5c5c04","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"23e817d25f4c411bbf4fca6fdb5af6b9db69bad4b9a452f403f60a6ec954392e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"a2dc9528e6fa3713ff8ff709159f17ca086f4af5546958f30661b5585ea9d756","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"371af8a05f2d85aaf49336f855698d5f05159f078e916a91f236bb9d3e62d85a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"49c6ae5ffc6d19c746859bf0c9e62f2844a3f2923a6135f66c483d542587511d","bytes":5114}},"requestSha256":"d0b7a086e07e19e30cbfd7fc7e343f48c3ceddbdbf51ab0eaa0dff17071b1109","metricsSha256":"5b47a21047b1cb898d14ab9acfc1757409b8eaf0dcecf90ab7c4061f155af821","continuationSha256":"f63011aaa78d82df7910294743cc3776edbd8de6847b46db9910e03064184867","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"371af8a05f2d85aaf49336f855698d5f05159f078e916a91f236bb9d3e62d85a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"1aae7f7e03b03f206d15bf7ac48d15f9d6b845e55f6bd00b0e5e1706f8fa5667","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5dd5fc2cc428fdf8761089f84a14dee1b29e32069a6700a9b190b58bd610b72f","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1e3b203729b323ce163b556cc93b9bf71859766b6d473f576585d51757e52de0","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2456281","recipe_identity":"1efd5fd8b5b8265ed3d73eddbe407e93e07b68afd944e9742c1faf29274944b9","loss":3.4647956002771982,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"c4c78c852054a3cd357bd84335b1c8f7586b89f218f32fa69ebe400bd5d1d355","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"6b8eae956d0166fe395a993beb6b150b5d3fffe19e8a220f03cb5653c69cad28","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"53d78580ac9adb1905fe986842926489660d09b6f2784f5011d37fe76d282c42","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e2a7de52f21aa0ac82f1342c47187eed5886130bfc4b076199a8fba736b806a6","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"9bafe8919ba2a58e72e7f24bb94a685c9d8b1be4332cdbef2e3a70bed8b86a3b","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"54ae7db12e73921a2638cf4d69aaf95b221ba0f60d69ab058607bfbc1fd5bdae","bytes":5169}},"requestSha256":"6b037f512f6b53e0e3512eae7b56e973befa4b5c47d8a73ad0ae14ec51a3e527","metricsSha256":"74da634eb6d4406ce112d98db8637c41ea076458caf1316e7484352d91cbfc14","continuationSha256":"e788e8abd033b426c47cb5d61bb827426c549165fc7619c34a6fd465daa28109","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"9bafe8919ba2a58e72e7f24bb94a685c9d8b1be4332cdbef2e3a70bed8b86a3b","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"ca2771eae583c8e715a366d997570b5f62e111fa08a0ab5fb23df4e0ed06641d","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"584e34cc76aa15b70badff5da687df642cfc1762cf9ee0253038378e5df887ba","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1efd5fd8b5b8265ed3d73eddbe407e93e07b68afd944e9742c1faf29274944b9","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2456282","recipe_identity":"6aa7d90527dd17afdb748b62d248cac692496cc0c06172f4b9eac09cf5a03d20","loss":3.4664831987501215,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"e1d5fd82225f437d0444e33d5de945e1e55349b22f58dd6905a7316e4f14a918","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"d999f45b76d6c64cf701afb563c41fa1054af4d8e677c9c63bd6f7a518061cca","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"968967bdb78646c7fb02f416372e25c8604b377a650889e48a4d8a771b6a6dd4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"3fbe59d932291a43868691aa80fb082713e1bee1817b132d8f363e74d016cd41","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"b7abd2c081ec2518790ed677132430d53063c94fdb02c4a3b432db7d68416a32","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"5c4dc7445eacc120668daf0180089d03e24678a6c1c01d12d818da45d91df863","bytes":5199}},"requestSha256":"46c594fe188ce1f25fde522272a2c5c0c5fde40d9984c27a0a0741c7eb58d6f0","metricsSha256":"12d39077317ff4073a02eef0b8d7d069714e0179f08ece0b63f86f04508d3ebe","continuationSha256":"e8810219e2a9dd85c8ee026b8a6a8917c2a9e218dff509dff8bb6299113a2288","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"b7abd2c081ec2518790ed677132430d53063c94fdb02c4a3b432db7d68416a32","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"59911485b62bd9ebdc43933b0cc798a50f30cb162b797bccaef858f71a637155","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b516738123645649eb78cc3175378ac7fbabd40c5a9bb118a3799d50946a217f","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6aa7d90527dd17afdb748b62d248cac692496cc0c06172f4b9eac09cf5a03d20","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2456283","recipe_identity":"24ddf086f12daaa72f6c8f5c97ea6dfe84a66a4da3c93b78b6dcc330d0097bde","loss":3.4613947454175875,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"2ff6590173d3043ae98637b543522f1107c6beb244b189d34865c5e77f5c1fa9","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"0b67a7c137440b3accb067423a90a21474f9083db4654f5ebe0d72afc5089435","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"d9076a9b4b19f4a6283fcab02dab41eac7b4ee7b2bbc793828949d1cc0fcd2bd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"93356c3aac1c40167a336d4bba0099c4eff989fa68ef744e94857edc09c9163c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"73c2415cd5ddb6b4019ccf5d849711d3458658f0113111b74c1fcbef2ec51cb9","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"fc11ecb87d2f6aeedb82d53f82d5f4a27063c1969464cc385399a76cd640ab2a","bytes":5114}},"requestSha256":"3c6e03e8826b6304c485ee28e7c00c14206e12f38781d95377740e11f4e6c3c2","metricsSha256":"1e641f6a74ccd8a80ee3fefbbf7e1a5949803b487e5811acad5f405b4dc91106","continuationSha256":"e459f7d2860202544b63c5142f3c25286fb67e0662b830f8602a5859fbab1e27","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"73c2415cd5ddb6b4019ccf5d849711d3458658f0113111b74c1fcbef2ec51cb9","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"9defdd9d2f25fa2c6c953e2e076f2506ce30f4a9c552610a4ae0fae92458b5d5","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"bbc5110848251cc53f8993ea9be4ce46edfd927e7ccdee4cc9e1fc84aefc11c1","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24ddf086f12daaa72f6c8f5c97ea6dfe84a66a4da3c93b78b6dcc330d0097bde","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2456284","recipe_identity":"adc23815adf8e2b13a725e575b8a7056b28165e2d61b4cdd7254f364fcc4a55d","loss":3.4676512611340207,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"2dfd9076287d0bbc45dac137e02a1012a134373a4f8c1e5e59bdd48d3192fea9","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"c0810dc54a8d2e10d33457cbda71275476871bfa1c89d35e3a36b650cad40bca","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"4682d9d3eb7e65648236aec67aac87a25ecc50c12ea65ab1a39f1e41d7782f61","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"b11c642f823e5e66258027ed0331f7916c0088f5d5a7d2958ac283ca1f11cf06","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"0118546d94eaa02101a4cb5353bde0c67b8812902451ad53b495179cbf65da21","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"7b18353b9ee168e4dc5e1b03a7119cc61766cba7a7ad6e977ec9e3853fc9df88","bytes":5171}},"requestSha256":"bc2555e234d4573264d4ff6dd944b070b5a85f3077d843e4957e20b17719c8e6","metricsSha256":"a46552fc5861b98e577059af49d13c4cb47444a1293f374dbda637a1162f4320","continuationSha256":"4fe93c1844610feed328baa32aeec04388597c766c1e304ddb7df83d663bbf58","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"0118546d94eaa02101a4cb5353bde0c67b8812902451ad53b495179cbf65da21","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"5f81341c457912229a7fa545629883dd25f85bdb621a03530cf11f8905bea064","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d0cccdc814580629bc2ec3b69fa2fe573ec3c07cbd04d49a5b0cbd56917c4d22","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"adc23815adf8e2b13a725e575b8a7056b28165e2d61b4cdd7254f364fcc4a55d","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2456285","recipe_identity":"1d033d95cb950b45c886b6d270b7a529af33b21475ca791ad575dca7418be2a0","loss":3.4697875636146684,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"1c1a0d3ca5c6fea8013d4202282c251b10147c864f8b9252db1bd367908e61b8","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"c2e436859f6414eccfbad0d0dfb495e1855f65f9b30d10e305c1fd0d9b455b0d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"d73b606845f108b70d355cf4a616ebad8d20bad0fe84a6da917c7fe359d66865","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"e9a65cdf3b8fb3a31f73c3f4eb06801650560ca5367ba5d39ea19b79db46cabd","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"4e8116845855962cba2573d53c696bb8f62dedf892d1112d7a702b336aeb6d98","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"21854094c679448aeeaa2fb9f6af1b6b5e17986fe9a2c256e5fb2db0e2a5d833","bytes":5113}},"requestSha256":"265686c88fc1e51391c361d459d5a9c4fcf57670e2998fc9b25418531eb41bad","metricsSha256":"698fdbecb5b9d30cbba82b934a84bc6029983b580a8d8b2f2e4cb618ba6c1392","continuationSha256":"255de5f8f50b488b2184813f8f08b35180d25f2ec113984c30310a6cc1606861","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"4e8116845855962cba2573d53c696bb8f62dedf892d1112d7a702b336aeb6d98","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"8afccf4e7f5f152c789bccff470c5802328341f1dda5b5b075b00081f004d81a","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c7a99c746d212c53e6e2dc86c00f3a44d116b13047529f88d2671db41da0adc0","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1d033d95cb950b45c886b6d270b7a529af33b21475ca791ad575dca7418be2a0","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2456286","recipe_identity":"9167733dbe45d80547253cc54cd9a7b1ba4f0facdf3f3727484d33a326c40d40","loss":3.4492329075080965,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"7fb2c12f4719802c446b615ec96f25ae8f014be1cfd47170cc971d2b7e697cc6","bytes":739},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"f53636ee1a9a3644b88af028135745d2b4ff1b5eeee8f509c914a95de2e72324","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"1685e4bad57f49fbd7dd413b46a6789b16ca2905299d70b9482923d596d2eb5c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"840d72d379b59ee0bed240a449dbd453c841b47d28bcfe4d7f04a258296bd831","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"7de1ad519e9204542e6688b6f36d5e614375a226d09e0a46edbed22ad15391c3","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"37241481cc992f33721c05acbdb0a321af641862c1e247e3556f5165bab50fae","bytes":5171}},"requestSha256":"2c88d8b1dcaa8a732d9041ca7b4c9d79d0664eea55e3973b1d74c418bafae455","metricsSha256":"e8d85f131f7071157c06ff854c0b6b801752fb06dff91c7b7764e891a5c4b186","continuationSha256":"59e46d57c5a262d85912306363cd738a544e015c2196c24f564d3bc157cf6e81","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"7de1ad519e9204542e6688b6f36d5e614375a226d09e0a46edbed22ad15391c3","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"1d4149e109fe19ae91fa77026be8c8e88458095dfc7a93339171a16c7aece166","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b9917fd47aa835a95becb81e8b319d1714035d318f9ce084d668252c8390a7e7","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9167733dbe45d80547253cc54cd9a7b1ba4f0facdf3f3727484d33a326c40d40","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2456287","recipe_identity":"c63c8ab822ca0bb7bec296eec0626fe054965c5940b6d9c3e2608c18b3431017","loss":3.4551217690214986,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"886cc05040dbb1b2ae4fff4ff82d16837a0a22d75b052ccde9cd24f6af0926cb","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"20a9d709060678e8fb2f0e96a52f01902cc42ce8e3d285efe6927b0e66600e15","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"b91bba1002b7bd39708527eb59aa2c01754dd0fa0cd52e7b196b6ad92b4ddca3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"39f7ce3164f7cb98dd0aa9d5298672ce890b35805a47f3250cd7f2df77e8e382","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"dd18be177db94f1fdb61c3b229bc83b40a164ae7ccee92ea3d4c9d96a5808d3d","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"8cbc83f20666dc1fa8747e485ff02cd1c79b79cb2d290ebe77bc709d7af622ca","bytes":5200}},"requestSha256":"b533f5ea84e3cfd3f1a59c30d5e8320f58533cbfad7702fe8cf855192b083399","metricsSha256":"756da355f9271d2c9b3c066843041e00ce7e3576e5fa9d15e84c16437c12903b","continuationSha256":"07c1bd527e7acd0679b45407ac0878fe9add26a9976aceab98abb910690c9ecc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"dd18be177db94f1fdb61c3b229bc83b40a164ae7ccee92ea3d4c9d96a5808d3d","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"aa1177a33f2f74f5c0f4aafc8507af72b2cd691de22fb3280c860f44ecc89842","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d7d0f17990cee0856327fd6bf5ab3e82806bfbaedcb5ae4040c33a1a701dd38e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c63c8ab822ca0bb7bec296eec0626fe054965c5940b6d9c3e2608c18b3431017","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2456288","recipe_identity":"8890d7ce98f1d566449596664442e1ff8e67813bb63dc1c358c6295fe180d24d","loss":3.4540458228008246,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"fbbb9c54ea91ddf480b43d93b82883206b0ce9b10a58dd039302a45a2e4969e0","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"81ba5475e181af3875673c2a5f6b9ac45201b1eebf0b4a3eeed5898ef1178cab","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"762e8adb411311b4185905727a7876a4aa618a2faac90a420e75a6c7859c3dac","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"802ab37dff09f54673d4713e41d627ad057bc7a53abd86ab15987eed9340fe8a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"ee1d7e6d43530e9d6562a408ea8c3003ad6d514b14913843fcbada4a41d03119","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"d9b3a3d94c9e2954dc22f3525d7790aa96ebe57c78eb86b0584801618fdc3efa","bytes":5173}},"requestSha256":"87217197d6f5e071636fcef566dcc394ef234c453538df7a0aedb925975f7ab8","metricsSha256":"66b607ef61492a6f549a2f0993abab11c8eeabae5635cd33f7ba7802d88765cb","continuationSha256":"d08dd781b4a97aa5406de3073b9633cd931e31707a33f5682d6c013acfc45012","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"ee1d7e6d43530e9d6562a408ea8c3003ad6d514b14913843fcbada4a41d03119","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"6c560d1afd81b36913011840853eab11d8814922817f9148af7e3666ef827cb9","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f47d72255d99d1c063cd9af85acc3fdd7f7b5ef6c4f5d626bc5928d0463506f6","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8890d7ce98f1d566449596664442e1ff8e67813bb63dc1c358c6295fe180d24d","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2456289","recipe_identity":"0d7722d5675bffbbbbbdf58303ce9a27b4c8e729eba2569ce8520d20a5119ecc","loss":3.4501401313581113,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"10d37367cadb90af59de4d23117540700cebd96a088da7b9cd164b5f999c7e00","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"d5a7495c891685b029a6237b3075fdedb924b31fc6bf8cd2424701c3b502b46d","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"de91de659425264f3c70a80b83b8c75981b1667b0817cf3fd36e89b57c7d6794","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"9a4c260d0c82cefdbb00ef08e53186e12194b48e6628653b2d3d25706681e20c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"ca9aed8d3076ded3de2f70a98530db3a4c92b8eec5203085f49196a08f48fe4d","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"a6f2d56ca7069abcbab61d327db1b89283a7af03b1d18af7e16fa7d8c881244b","bytes":5173}},"requestSha256":"94786b015146e2ba27c99b07ee1519bc761857dbf4cd6da77e03e056d3a29646","metricsSha256":"e29d804cc285cf4d4f362e44307f0536dafb55d58750fec1bd6ab4d6e803a4c1","continuationSha256":"8cf182968e0ddaee745fa61419314de3b671e9f158e5eefffb12225119478a3e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"ca9aed8d3076ded3de2f70a98530db3a4c92b8eec5203085f49196a08f48fe4d","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"1a8f99fc669638af1fc77223aff6603f56e21f4c9952715ec247666bf1493178","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3806f8012353f95b959acf9f17d705bbb4d0d7d80389974fc4fbffb9377f2514","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0d7722d5675bffbbbbbdf58303ce9a27b4c8e729eba2569ce8520d20a5119ecc","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2456290","recipe_identity":"b91d431e5dba0df6f9c35fcb7ce1bcff74510f55ea7379068ff3da70acc940f8","loss":3.454050054316837,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"bfb34c518e7aaf2353fe99ee0d3122f0b61fa48fb1256f30d1fbdbded3e39440","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"f6c9688077c024834ce9675d39e00c95dc5e84868c796f9fd54eefaca05d5a6f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"3cb95b02ec4a217009942fb809a2d5f17fd92e5df82d6c25e8d33b43d982624e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"16a2c0c4cdb5a751378bfe2724398fa978b1a89065f85aefd8cf073db7e02393","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"a695d78f11f98e7ac1aa0ebb50b23491db28a48fa270bc0841e86390927882d5","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"3572e8cb245c47d6e8a44e4eac9bad850ceccd00aa7861fbd8e14dbdab48dd58","bytes":5202}},"requestSha256":"c8ed84e5ca92823161e4b92921280ddb7139697769766917b9de3ee26e68cefa","metricsSha256":"d7187cf131fe60ab51c57c71755cb6028ad1fa1a33187b5c6c931e3b13ee1059","continuationSha256":"2fac47e739a418fdc61fe50f68b4c21208abd63038aaca7c639985937f1972dd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"a695d78f11f98e7ac1aa0ebb50b23491db28a48fa270bc0841e86390927882d5","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"863c68d74212425085b1f46518ddbd49f39b9d6945d1aba5777d6caaa18e4689","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"bc727f09b952ac9ff11ecda9173bde1fb5d21cde47597b56424071a8e05ea15f","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b91d431e5dba0df6f9c35fcb7ce1bcff74510f55ea7379068ff3da70acc940f8","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2456291","recipe_identity":"51f143dfb4573d23b4e7809e94e281c98bd55d8f3083d4e2c4bc77d7af0755e6","loss":3.4556786315073427,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"22cbd92512f27fa9d70b7c82e35a1471119a564bf61ddccaced9d6550bac96be","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"f238fbcfa166967b98bf995e3d9d9b981907b3e1b29552dae0c680d336069900","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"e2e8ae1c41981dc541f8fd8ce026c7d78fe87bb968c9c7eaedd7317930de29fb","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"ba5fae78a26658a0759f1ac0dfba0a7ae31cc1438a4fe8539ac298bee627d10a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"6e13e814ede8ca438628d45b2251ae7b3ffed619a8c6c04c0f53663250ed1cde","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"632f7bd4ea22adb9f3cf227014b64a71312c4004f185e6e64db0f2a313c8789f","bytes":5196}},"requestSha256":"89d5aa21992134a6e4d12ae3e5c9215707748a061880575e150562fb6c0d6500","metricsSha256":"68ce5ed7c30464417f8b7506cbb76019dbbf664dcf92f2ef17306c4e8f1f1ff3","continuationSha256":"5c4e2c43bb75dda0e2576f36ad5bd61e4948e318751e2c9610c4e1d28e5a9842","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"6e13e814ede8ca438628d45b2251ae7b3ffed619a8c6c04c0f53663250ed1cde","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"fd59e7e161bb4b78eec6e5e15e3df0c65723afd466eb761355b7fd140f7dd2d2","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5684cc969d18e7a8638015b7bc30cf72355ba58c28cc373c0ab30844c5bb298a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"51f143dfb4573d23b4e7809e94e281c98bd55d8f3083d4e2c4bc77d7af0755e6","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2456292","recipe_identity":"e49d6a19595dad1d4fc245b8fdbb0779d4e8718ad0bc3745e29cba5ed0c711c5","loss":3.453656531049864,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"cdc5dbba9b53a76075dcf22b017ae1a5bbdae42fd0d0743b94cd99327d34dbfe","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"904d8ba083c3f9193bc7c8b5a528c070b5ddc5776086f437743ca63657d37c44","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"eec769add15dd43931d5abfa6bf81664a870b4cb556fd218d16e8d224cae8d3d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"90bd0736d096dd3757fba19c3206bec340a01091eff396508755081388fb55ce","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"fa8a5cc637da059f088a96704ffc37b7a57e40e1ad7ef4bf209aade58bdbea33","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"e592944b739eee095690692914305ff395008244d84bfd986461a0bc3973b85f","bytes":5171}},"requestSha256":"030746f0099b4bf82c8eada328dafe469764fb8dab9ea0b0220955f26795b954","metricsSha256":"9ac7bfa0f6001e4d19539dd62b8115b547785bc9fd62fbc571d92404fde9978a","continuationSha256":"97d1424897c25cc3b8217d71414d66e2f3d6e338753acfbdd1fe819f0aa11c9a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"fa8a5cc637da059f088a96704ffc37b7a57e40e1ad7ef4bf209aade58bdbea33","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"7337b07e987a54bb97d7d97de610c48d1e7c5cc83b727d28fbf0026803682b59","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"00f7c46b70aab600be222a583e855cbf1235ab9f33b0d3e7a1ca4ab2957723b2","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e49d6a19595dad1d4fc245b8fdbb0779d4e8718ad0bc3745e29cba5ed0c711c5","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2456293","recipe_identity":"98ff93c4852447fae32a4bea110beddaabb15c63e0da884fbf217d8cf0547a9d","loss":3.464871085681808,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"9778b0610f40c89934994df1f298ead9e278b1b884c03baba90cb38fe0e2d0c4","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"90aeb8ae2ab595bc6c62b919519e056601b54fab47d6ebd945ddfb5eef75e2f5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"8f6677e916b1ee1799db1987a0b87443ba815f4fb8773d914070bbfc7dab8446","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"3ef02c771e5a53609dc82d1e273912cbdf4ff23bd2f8413507e1282511480b17","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"f803514e5f692e5f477351a7200146210deb428cec7e4d2804ce3331e77f77c9","bytes":244949995},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"4fe53a0f5ceee07b971b2837a362ed325057de8bc5e5edaf4433cae8cd8d1300","bytes":5200}},"requestSha256":"00728aec9aa7ae01818652fa355a50ea760dfd839e296015e1f11a3d40ae663b","metricsSha256":"07bde58cecf97b2707baef6f57260c06abfc429bbbb53cbf89e7df72464191e4","continuationSha256":"28cff60284e909f0c6b904d07c2561a9efb70a2ae2e00f407868242bc0053e82","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"f803514e5f692e5f477351a7200146210deb428cec7e4d2804ce3331e77f77c9","bytes":244949995},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"7a82bee2066f1c2015e93b7cbafa1dc0a53b53612b671685acbb327fffced51d","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1ffa38904f628ed9d48d0eecc32086871441d0c2609b959096f5cd2bfd206838","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"98ff93c4852447fae32a4bea110beddaabb15c63e0da884fbf217d8cf0547a9d","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2456294","recipe_identity":"fc2e9d5bfe6c344df81d884bcdc8e7ebe0556cddcac7c18b2408ca1b0630167c","loss":3.45512636877735,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"eb0a0198d52882a01cecffdd4bf01cce60d451e6f24969228eff045ff32e23d5","bytes":741},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"192c074f4af1042a91ac3803f2e09837ca44b887b55405e270997f8b3b5b27e6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"a637498627d7f5604dea94b7a0fe9e70c8121888e7f8ecb2393ac511f6cb4214","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"abda07640ed5985d26ee798f30d8396b50a198f28ace1e8696d4b75c962b1632","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"b0b68a407bea7816f6b675b474528269734b1dac72e00f6344e67d27cf66f56d","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"6274005736bbd05af45b5e9261f6b16509dda946e95680b5bcf679148c98638e","bytes":5200}},"requestSha256":"088bbe269f6a8dfc68edb231f811d83516c8c1fee0be21fca5f45e97298dfcc5","metricsSha256":"db9d2892a057d29dbc220f3d71b69bad3398c329dc4cc8105171e68fe306880a","continuationSha256":"dc81d033bfefa7591173a34b4af7fb2eb5470abd0cfb0c0cb0e5e4a32bf31888","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"b0b68a407bea7816f6b675b474528269734b1dac72e00f6344e67d27cf66f56d","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"004da58d6d07765a53fc8bfa65e84997c7924786ecf65afcdcbef85c27a1eff2","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"916bf5b877d6d81347b1922c7a4c7180398da5f1bc34b4a7acd01dfa50a4f5ee","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fc2e9d5bfe6c344df81d884bcdc8e7ebe0556cddcac7c18b2408ca1b0630167c","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2456295","recipe_identity":"e6a22cfe38823c9e46d2095a18739e777ff36b0ec1ca27289197e07c069c9007","loss":3.4520953789037487,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"e3a5486f8ddc9b4251ee50f94bf69fc5a82d48a9b3e07624061864f3e3162049","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"30e992e7920b7411b9f2671dcf5de17c6f0572139879097d898f2ffe530fb4e8","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"ec16e4acc24e67ff03ebc0a06351cfb6fc341c7ed70d354ddaa5fccd9617ff22","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"81be75f860e5322f3278d77f8b2e621f69f1e06b15bef1f3bb7540c4abcc0d66","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"3ba0776950f5d73be0b3d3a79097c5282c683fbe246dc74b2acb9c605132f40e","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"7464f2cc931612edaa3832b6121563758614b97f321bb8e46fc8991f91e42dc4","bytes":5114}},"requestSha256":"a69bc4c0e0c10115c56286e1d8db8b7d91bf14394d582974504c6645f0278932","metricsSha256":"6369e7ebd4145bdfc6927971e4abb796f75502cecd1fde872bdfbda7c9784e9e","continuationSha256":"c6376a5802d9a11a6f4cdc92a152f950793c7dff95808fecd5a204a9b171eb00","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"3ba0776950f5d73be0b3d3a79097c5282c683fbe246dc74b2acb9c605132f40e","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"e2b5a35c9190e1ad16b3875a31f392ff2a82f08a02aa8bd3692145e876133e2c","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"7a9def965aa676ddaebc4af68bfc07c3d891886cc1a5220273a310e1ee41a002","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e6a22cfe38823c9e46d2095a18739e777ff36b0ec1ca27289197e07c069c9007","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2456296","recipe_identity":"29f9dd585bca19eb140978c692c59850b239137ed186a41202361921b4532c25","loss":3.4580368830143953,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"a17794d951d4ee61fccbb9a82c33df95a08844f6d5c21dacb31440fb4b649f26","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"a39e44dfd31039985adf3458bd2ae103d0533b785e59184e5fb8065c7798af10","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"74b0836ceca1ebdbdc2abeb71b1a670e514ff59ef9decb0608ec69fbdd37242f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"5aae3374f5323f0c66681cba87dd5447ede150183affa2cbe305d572e32afaa8","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"20eb80f0605eed7a2290120d04fb40c15413796ec0ed75ac129adac03dd4e6a2","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"78ea6f33544b8688713d180b7d0a3ab6d6b337995fcd71d66ded2f191360c89e","bytes":5171}},"requestSha256":"fb90641bdc3252360067bed3a29a33c6921f2f2a8fa97fefaa155fde135802a1","metricsSha256":"36447cae2d4ef44c1b17211a8c5e92066ea1fb1b692c541e89c7825e623eca3a","continuationSha256":"ce89c30be9ea1952dca96e38442a968d5eb8576c5882c822133a3ef4c31fd094","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"20eb80f0605eed7a2290120d04fb40c15413796ec0ed75ac129adac03dd4e6a2","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"dac5c46dc2e5add42ba9a43a9c4ce6a40e99eb6119ed348821d8c272985ba618","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1839067f3a053254a27d3fe93aebaaa467b0c97b8ba7c19bc95d00f72311489b","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"29f9dd585bca19eb140978c692c59850b239137ed186a41202361921b4532c25","loader.blocks":797056}}},"sync-wsd-h40--0.004--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2456297","recipe_identity":"040a82ca6726124539b62558deb1a4303b18c050a7f42c4c96ea09ac768aed16","loss":3.4554055779772432,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"900b7aa22e42972f6a6923bcf39892cafaa9c0cff3d5132b1bca2c541ea2e5b0","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"1d82b7a1959ebea754d75c68200114f07d9375e3472e307d694d58350f94c0df","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"29e764f4b392e639a3972ad83c21de572d33f18babcf943ff076a204f3a6d80e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"51b632c5a090b4e2114e497f3d973a71c52866d14d080269981c3b7a200da87d","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"9cd1f82ddf28ac0fbd24686bc9ada19a2556926979a463068cfbcbfda7992fdf","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"11da03940242451840f83ce028daa819820b3cd9710f193dd31e982ec4fac5a1","bytes":5200}},"requestSha256":"9cdb3c7f17a475f16bf6584adcf94364d943d68cd4fc2d96b021e8b297f13649","metricsSha256":"625ebab988a0b39de158c7bc7f4b4f7cdf2732bc870e369ba800516075007ca3","continuationSha256":"d36ec44c464f0762e47d94744a2593cdfad02e84a51f34b710a2e85a96b1c889","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"9cd1f82ddf28ac0fbd24686bc9ada19a2556926979a463068cfbcbfda7992fdf","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"339989ae142bc9a49faac15bcc43bd2f4b57e5146bcf45258ff6aa145ba0dfb4","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"012b311a4fbb439c0aa5f9383dc99ceed882ddc80e03145aa2aac5b23f4ce6ce","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"040a82ca6726124539b62558deb1a4303b18c050a7f42c4c96ea09ac768aed16","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2457298","recipe_identity":"24cf9859fa78e12ea697974d6c43abd0570001d5926536c2e4da988c774c341a","loss":3.452309710920177,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"22f6216ce9e8a3946da67809dbd533fea32e3ce4a5462e0f23411d2fc34b9a8d","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"6481c25eb4e134cc6daf067e804efa6c6e013157d52cedc56f2383db8746a2b7","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"c09cee61d3e42783b120ba34754ad27b5aaa21f039347e16673fe9b35cfa5d58","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"65008a2d5bd1ebced6db7ca21b49ded7ea607e2119e81e19edc1566110a213cb","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"99119f3f75eda94fc83663232ec673c93f5037986984ebf514cc2d640af65e77","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"babfdef47f3053e715c1b385f37296b925ec20a45d6f73c1de96480510a84316","bytes":5168}},"requestSha256":"0bb5e3131c4f6703e8cc961bfa8b53eb76fd1ee00b0328d20e699338e1433575","metricsSha256":"6516371db09ca4772b9bf8b22774753f168fa9bc15fe1547db56de94f1367096","continuationSha256":"4095015e6e1ca6b7b3b059d3ec538bad84d5cd4628c707391f81fb07c19ec582","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"99119f3f75eda94fc83663232ec673c93f5037986984ebf514cc2d640af65e77","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"6a52bc8638f3744afed53000e37eed19ed74e4bdcd4a58217aa9e41dd7737281","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"32b8adddfc7344bb26b90362c7c9a892c4e4270380db2e0f83469989e35e9d76","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24cf9859fa78e12ea697974d6c43abd0570001d5926536c2e4da988c774c341a","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2457299","recipe_identity":"0584d7e3d6e23d96e4089284e7b144ed42e49ca3c66cf4b4ac21e73f79008ee0","loss":3.455872899370765,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"e80ec7fe500919a41c55b4afb9bfca571179b41dccee8f547d3e7771d5b5fc9e","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"651928de31a9b454ff3b6a17d6056b942665c1a0e73050bcc8f5d74c7bd03c20","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"9698f847085d8c5073e08765889c14cc5659a93e46db6769ed7c104b521db985","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e9a100e43e947c9b1042b7ed914769b7ff47ee90fc609fe27d00d77d77e834c1","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"ad2652a0c097e7843d0639686d200693e7d2f190f2c64959c90eff209e602062","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"565b889dca87cccbd84138b7fdb9509d9f5505433ea962a11026cef2f7b43b3f","bytes":5196}},"requestSha256":"62e1205ad78b42aab7d4e1740ddd7b70ffb03a1f2c6ce1708e542259bd4a5ad5","metricsSha256":"076f28e6780542eda91935efbb6ac708fb9acea1bdcac0923dacd24d639eb849","continuationSha256":"1ec4266a91fae8c629d9d93331c3e1e0e6c948bb9eb6c7e9dec98259f1e37066","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"ad2652a0c097e7843d0639686d200693e7d2f190f2c64959c90eff209e602062","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"a04690537e3a3bc292d92ef1a2b166231ec93d67ba666d162acda8d58dd8bf65","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6472bcff2340dab0523131708cf3d0caed47268831ab992701149e32eb39dad4","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0584d7e3d6e23d96e4089284e7b144ed42e49ca3c66cf4b4ac21e73f79008ee0","loader.blocks":797056}}},"sync-wsd-h40--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2457300","recipe_identity":"23c73841b0d9e7d9869faf5dc75f6557a62959ff4161478fb981742d2573f48b","loss":3.455832290668222,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"eb18a90998df2f8fe08ec20cec04437cdb10e089d0977124c67c87c325621a71","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"ffee8dcf3fbd76f6f011bcabd95e881836e79c2947915c4ea18fc8b91c69dbab","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"5fd867f40865d55a7305230c6a62cd3bc5a3e0b98722282ad4c2361222a8e3a7","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"9f7bc56bb57d9aef4d4c2d633df971837f7b514b8f4736506b7190f8cb77791a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"0951b0e0cb6bf38741548208c74c10df7d7d7a066a742948e52b0a674b7b8de5","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"54639f3e9645a8c94741fa53a834a85d1de0679e1a28b712d79b2de5d52e41d7","bytes":5202}},"requestSha256":"9d44cdbd16b6b9280817447425da1d5bb487f33bdca75493e5816c0ab60ef488","metricsSha256":"0058f5213115d51232b09f0583404dcf039662dd35396f53682f3284f4987d93","continuationSha256":"b0a34a4cf9942fee0b0e7a17c71d765ae2ac5522573ce91edfe2a8cc955d7575","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"0951b0e0cb6bf38741548208c74c10df7d7d7a066a742948e52b0a674b7b8de5","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"5dbdd3818df177acc936a5c65de0e14bb93e50daf84878f6e57084cae1634d8a","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"dee49cf1f9785a4bf1d721580686ac9814141590e8282877ca95703b3dd8842b","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"23c73841b0d9e7d9869faf5dc75f6557a62959ff4161478fb981742d2573f48b","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2457301","recipe_identity":"663afa44c4a430a0efaf6c95b14587475e07dd6d71cf66248e917090cdc2db12","loss":3.4549467369817295,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"ab216bd40ea0c5eb018a94005bfe50ef73dfe8766f561404eccc50761cfe7b40","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"efcba51439afa0aedbb56762133e3a453f6e79a53500c51052dea559abad6800","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"dcc678e9d8277a5653a2a9f72fef809391a715a8de9a1756437fdc96f9a86737","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"eceaa36844461e2478999b3f1436a3ea40a907b41faf9607bb765a3f575e55dc","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"89f1b1e6418fcfd72d0e6be546ad0eab7d376b1c86f3897a0ca9087380e42929","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"ea55db93bcd6d550cf1c9509a08381ddf8cc78fd82206aa27da43d5235e3e774","bytes":5116}},"requestSha256":"5dd784ac8b36524a83ae2b42458a822bf97a2db237c6ec013b3aed71a5445824","metricsSha256":"5728e8da89341a2198524c913926ec9a6f20386211372bb10cdc21310eb5b0b7","continuationSha256":"dbb6ef2af85bfde9a4d87bf95dae2d7ac67e4a2a5df232d9399703976a3999bb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"89f1b1e6418fcfd72d0e6be546ad0eab7d376b1c86f3897a0ca9087380e42929","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"b83e3953937edff56c0985850031a9fc2c699eb668670ed8e2894026db5fd672","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9faa0f5892a72d47793cfc6cdf09077681c9d681ef1b4f0717930aa2a1518b49","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"663afa44c4a430a0efaf6c95b14587475e07dd6d71cf66248e917090cdc2db12","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2457302","recipe_identity":"ceecd345386516d1361ae215c344b1c839b14a0c465dea3931948ff59b0676dc","loss":3.4634011282309762,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"df9fa87ea766f8f59ae1ac0f5fe31f8651639308875bfadc3f2841fcd939d03a","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"99db88c08f0a73e8f8496bf0ef2b6f4808371c3bce09ec82bb0c1ee91f0e0c79","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"77e02cea8e7a25dbee235fff64403b3d716bae7020d4ef0a330dd7ea5bf52756","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"220fa8683234fc1d63dd8a06b1dbf23a22514805f6f326cc40dcdeb78add38c9","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/epoch-035.pt","sha256":"170dd6233ce361653789c322916b848e96816518b39c01516ffe9d032300bf61","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"fcd384a057461e28aaacfe5065fb3e45b7ff34aaa4243c09a59c41e1c6035b1e","bytes":5202}},"requestSha256":"a5c4615a27ed3329de1c73e65d4e8bd1b8702dbe33535d9cc1385ee935c99a6d","metricsSha256":"42d8e6f9c2d227b6a6220f092e7ada283e7948dc4a98a20855fb572c362cc909","continuationSha256":"7f13afbfb9c2c5e847549313bc1e4ee5cc6d005cbf0cce07ffe0ea6dcc2d4bac","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002/epoch-035.pt","sha256":"170dd6233ce361653789c322916b848e96816518b39c01516ffe9d032300bf61","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"1ba17a1d8017b2a90b01d03fb93f39f2db67297237342bbfd0e939e318f45830","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-002","cpu_threads":8},"decentralized":null},"recipe_identity":"959f20efd4d07b81c4ff1d15147b0466e5d7457c110e7d9cdb764d88703367d3","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ceecd345386516d1361ae215c344b1c839b14a0c465dea3931948ff59b0676dc","loader.blocks":797056}}},"sync-wsd-h40--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2457303","recipe_identity":"8fc3a65934a1ff2f23e2f2a3f23b262daa7d3faa5d8da474b746544dd2e186e8","loss":3.4576668361233858,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"4ee0addacb0b514d1ffeea8cee76932bfaf0b951ec024fc32ad6a3a350356172","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"ba92b00ce7b1dfaad2ecc17adeaa446d0941f4af1d62cf2a50b7fb89e89ef86a","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"954294cbd25452e7ccb9c7bc0468926f2c33101ea49ec4263a032dc55f634b8d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"50fe7708535b4d926cc966939715f62abab8ab01391c6bc626977cc36b03e099","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"aa1bd5df55112d11df3e7176b71c0c0e31f4deca369c45fccaea7d37db6691dd","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"445fb4062358355c4178a35c0cea4d82a9feb4031f1af3b0ef995a8815fba011","bytes":5201}},"requestSha256":"af7e2579e41f08278989d3a614864e916eeec9818eccd67640af1a0a8e97467f","metricsSha256":"428ef8cbc9cf5699703879f8edab3693611c46fc9296e87ef5db4d39ec94fae0","continuationSha256":"ecbd1b3137d1d95ab8c90e638b764f01c52d09141352073303e2c6294bb8c6d2","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"aa1bd5df55112d11df3e7176b71c0c0e31f4deca369c45fccaea7d37db6691dd","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"ea81009a321815cb1838b2c3ffbf9097bf1ea43581c9a597dc7059895b7bc7a6","bytes":244946027},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"07922fc5b47e3f7e2964e4c181fccec5ce2e575fabba8a8497e1657dee05e8bb","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8fc3a65934a1ff2f23e2f2a3f23b262daa7d3faa5d8da474b746544dd2e186e8","loader.blocks":797056}}},"sync-wsd-h80--0.004--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2458992","recipe_identity":"f740e816691f99767dc6700ec3ccbcfdc10205a4de19993549875899ef302535","loss":3.400883827208692,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"424228d54d55fd3c3555083e009f35244f1f9142b1d0ed0e9231ce75cec7b832","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"a25eb88fa0b879abdd8badac5e8e9ead1fc995b72656cac9014ac495aa35eb15","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"66cd142084764143826d9848c129d635e371358470843b67b5df38a92cc7fc4f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"78830f6ed6515e503804d8e626578df7d43a727b6b82be3d9837ffa7b4e96e34","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"f6297ca4bee09ccecac8a8268250bd6e671f178b9d779df78c1575c604f4bc4c","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"109bded2f8742b2884c3d8ca027a7172046c54080cb51b73d7f468826e9a830a","bytes":5198}},"requestSha256":"aef15ccf2f12aa27c8d31a2ba641b89dd4230a7ebdeb9005e3f44d410942f5ef","metricsSha256":"e5d52ec745ccd0c44d432c183b4c303d87f279a6147d29c5688cd4685849a47b","continuationSha256":"ec8ee8b489bc604f0476d3b20a6e2a0e82611d75e1666d11ba0727e67bd89fe8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"f6297ca4bee09ccecac8a8268250bd6e671f178b9d779df78c1575c604f4bc4c","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"4ebc814dc747431adf8b6491cf7de47aa1edda1dfe0983f5a886bcf306163ba8","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"302c51f95f86bf4f3afae463701ed00417f7ed6082456399f91ddd3d50266493","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f740e816691f99767dc6700ec3ccbcfdc10205a4de19993549875899ef302535","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2458993","recipe_identity":"367af306003c82095c639f04a380ed4f7cefb1348959e5eb3248e843e9df3cfe","loss":3.404250131254967,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"7adc748856b30448d4d2aeefc4f0a5ff0b2090f031d92fc2267b233986378217","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"bb8f40c4137988fe383aec8b7f3495161bd3815e3152417252ad833cdf1b3823","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"be761211985d5ba8ad688e8df8aa990a9750f131088b233f2e78ee9774944348","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"c53155eaf5635d4af725522aff55b8feec5f42f435891e1a0b548c53a019f2b2","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"d20d849f8e4225d9b9ceb150e0f0052bd549bb6b2c9e3a0ade8f16a1c531396c","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"9a80d8c6632a349b31152c5501ec70213bd8a7c9a879cc28f57166494fc7a2f9","bytes":5169}},"requestSha256":"2d82579a8ea23bb195c754c9455dbb1d0d28c04a30357608fbfe279a2115088a","metricsSha256":"72b7fe7cf9da8bf9d8aedb18174ef82ee23425f56ce89a17a0de27520bc620b1","continuationSha256":"5ca95da273df2c020d3c7cac9a74f8faab416b6c6aeefab32790fc861977a0bd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"d20d849f8e4225d9b9ceb150e0f0052bd549bb6b2c9e3a0ade8f16a1c531396c","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"ae1e011130ad5f7ddd5ef2425148063cb929d4d963f3c1d18b084f2089120057","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d639aaadf01a6c3fdf9501925d52e937a7f30f2de708c7404ece5a0c6340bf72","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"367af306003c82095c639f04a380ed4f7cefb1348959e5eb3248e843e9df3cfe","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2458994","recipe_identity":"15b8c78ab97a784a7876c4c09c314ee7ed3707cd92a7291633d75ef50eb826f9","loss":3.4025661407826067,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"3066860ddab06a9f1f81836bd1e6fface4aee6a797fdfda57f28f4cfc6b7786a","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"ea658d20f253493282b5992f24ace6e2dd268314c5ccc79c7600134dd6e7aab0","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"9c60d3666a8198326f9f1e335cd666f205cef8085d6f7bee8bab5656d6eb7742","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"348aed42a643ba04e2e38371d0e1f5acd7aa0ab6e774a74f0f1881fade8e51c6","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"cb496f2e1a09f6a8bdfa8ba02388ec9f1266688883fa025b0a791464dd17acb7","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"88b087371ccd214d3e86b14c3b50d72665183db8ab6d7bd264dedb25833ff010","bytes":5113}},"requestSha256":"c12e7c8620e3d4d62359dc6871b35ba9a2edfd1dd3ac0b2e4e4e912909fd6c3f","metricsSha256":"fcb9fa88c2c6631040b095d6241a2f6348a5877f933b54d864e1bd3600ba9839","continuationSha256":"44ee7a47a7ffa4fe5de85ff627e1f003d2a81a0fefa4e0d5d461bb112aeb11b9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"cb496f2e1a09f6a8bdfa8ba02388ec9f1266688883fa025b0a791464dd17acb7","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"dd2951466ff7cc5fc6f0a1de999814630b749cad4c67950c412df171b83d5786","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f936f71bc4dc4210dc4d132b253b4a2ef3ae12986fb1d05de8133806e4d9e5eb","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"15b8c78ab97a784a7876c4c09c314ee7ed3707cd92a7291633d75ef50eb826f9","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2458995","recipe_identity":"1cd13860e821c339a3e07df7e607bc3ef38a7afdb1f67850da89dd516d72b81e","loss":3.406432859902007,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"92d34af196d5a8544cfc6c4957aa8ce4bf9cda05000a40d562b87f497d9809bb","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"44ee36b7b66f99bfcf312965fb2333da22ac3019ca3885af37788fe2d8e333bb","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"8b621376c3d7f0e2e5ade3cd58abc7fcb709f14c02e1b2bd3caca5fba667caef","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"9b3bb840968c1076e4845e07fc328ae7e8a4563f82c962754a0d6f3e59eb694f","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"f8786490c314af7f3b9ee2a3985a2c7751012fdcb48cdcb93f6336d4303bb33e","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"795aeabcaed1fa7e8c928cf18d7e134a0c845028fd79247783312ea808cebcca","bytes":5114}},"requestSha256":"d6235e0f9dabc2580c35a3a0cc09d49d709b6cbd327f8156883a382f819d27c7","metricsSha256":"3d64cfba2c28f4597fad22f20f1228dd1e284b293f6ea186307d242be8a31042","continuationSha256":"373af6a1d95d45c7d15d371c751d1d7d37d47d347ea4ded9f0164018957c3d9c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"f8786490c314af7f3b9ee2a3985a2c7751012fdcb48cdcb93f6336d4303bb33e","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"10bd837f44dca21c57b7bd7356265759c6be78a1e4c4690e4af614fefffb4819","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"4b12e59aa6a342f251b89ef63ead8a8e1ba5e92cf88d70b6db8346d68137769c","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1cd13860e821c339a3e07df7e607bc3ef38a7afdb1f67850da89dd516d72b81e","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2458996","recipe_identity":"43ce31887392f962466387fb450cfbde347d49c252662cfe1e74400ef3a115f5","loss":3.4009907702655466,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"fd3889e136107321ae71730091d331bb3e3b3298ed37aad34528526e5ca67b69","bytes":749},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"53bf94470cc772350c4ef57bfe5837d40111c7b928245c0e87b0dc02e99492e9","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"53d9d0ba6b6c23ac5418c60543487790618315bd7c532df6316ef2b194da44cc","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"f1d2ea6c73ff0363bd257eadca3912d97ee481cf0885ab16b06002cfd568ea8a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"516e6d80190f739934bda93cad5ebfd702243acfc00f77c7a371d8fb592676a1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"8f1503e59f6532a766d6ff0b09db5150ef6076f6fa63526f5ea43212b286e740","bytes":5200}},"requestSha256":"10349019ea8ae8009736add455713b492ae001a2c83f2b4488ebec27bd3ffd6a","metricsSha256":"389fd9ca898d620e7225861fb3dea68e9d4232e439c949ccd62c3505fd9c435d","continuationSha256":"80c11d688db361fa398051b8fbf38c28044c2d730154e48a50ce4665178039e6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"516e6d80190f739934bda93cad5ebfd702243acfc00f77c7a371d8fb592676a1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"675c6eb73649a41c88441d093e8d0c51d90d2b19dc1edd6c0856ec587842dbbb","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c8132a16f74661dd98af3f377f51f5a9bb4d1a7f343d95ded98a5d8c515505f8","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"43ce31887392f962466387fb450cfbde347d49c252662cfe1e74400ef3a115f5","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2458997","recipe_identity":"25f691cbff5df50ce00e22d786622d75a00855879a66c803c0e580ad37d3daa4","loss":3.406536946355269,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"e90a8f1d481b9b259250521e5dd6844eb154c7375b944dbb228e969e1976c25f","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"3dfd2e1b2fd53ee0a5d8e1f9e9fd945e880a80e4752dc6db5b92d0a3fc1e9ca6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"d3c587732ae623a03dbd517b5a91a5977b7b1839150d9dbb03d05abec60b369b","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"52cb6e9eef4d49cd9b5a10f398aa0eac86ea3da33d0737c5706db380fb099da4","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"5e5a879f1d606e95afb158837b3952e296c08bda8503f78ebc6415c317ce7e8e","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"beeae2ecec755bef23668110fe78bdfb1178920d5d5b8a6393ef0099e629ad4a","bytes":5171}},"requestSha256":"699a5aebeef01ba7e8784b7234d4bb70a05fed814cc7d81b7530e50c7df524c9","metricsSha256":"332441ca64078d227919c877203f93e4d20c393110e27183308133cee5d6dc73","continuationSha256":"d9d6f41cba00195dd1f71e2c57414cf87ac2c4971f00c7679f3010a7d36fc546","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"5e5a879f1d606e95afb158837b3952e296c08bda8503f78ebc6415c317ce7e8e","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"787f8dbb1502dc99969969c7e96e6696a45e740c8ed1e1278d5e5f4b8d29456d","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8b583b6c0fe88384b694511e6bef89c5ce1fdc7815e8e22fe4386a6233792596","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"25f691cbff5df50ce00e22d786622d75a00855879a66c803c0e580ad37d3daa4","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2458998","recipe_identity":"c19be9d22819877d62c65ca64121218cd08372ce460640572eae2f527caa894b","loss":3.4138426364204353,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"12c713ad31b1f241cfdaa1c71d2dd06d94bec85f6307744d19c646ad47c81ee0","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"0b1e41235dd4d4c09fe57b1c2338b9caf8dbabcf2d9ff72b6a676090ef92cf41","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"980ab88a3b8e1d58807aa3ca808336054b217ebb80079d8e2eff5ae6a155ef13","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"2528475b0d296edfd218d95c03166d6919e986871ce9fc972e187df75451e937","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"bd01f17232a927c208cea0ad427a058be11eadeb637506383e9d43f81c0dc0d0","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"7595ef3ceb7614942fb9c70f9c44d7ec4d4af087344b293052787385b83914a8","bytes":5171}},"requestSha256":"0805b9664bc95400fbacf12a323be14cb57ed2b4461d5ff51193eab474f5fbd2","metricsSha256":"0de579f89dcc26e0e1d9ba29499607505d86579e6ec7bc191fa50cad80df096c","continuationSha256":"2767fb057509a1cb5902d190865aa23ffc2a1a3589887c3a0d331c09c2a1ab28","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"bd01f17232a927c208cea0ad427a058be11eadeb637506383e9d43f81c0dc0d0","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"d58a7aebf1ad61c6591ac125e50c37a15e6a86f0966dafd452e0dbc2fd9e8972","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"60b6f06523e461bddefbf4847990b2724cc01024a051634648a79d558a74dbd3","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c19be9d22819877d62c65ca64121218cd08372ce460640572eae2f527caa894b","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2458999","recipe_identity":"3b318251c23eac7a41e111b12a20e722cb594442edaeeaa3d62659832b66a6b0","loss":3.404958759804277,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"9e8ed3e95d171953569877b8c8215ca80a7a7bd77fc2a8b777348c140fac7045","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"9635b4ac4f317a3d24ff414fa0488c8c5647dd59871d7c64baa024315d03494d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"0c7062e9af304a5278df06d4e168eafce29934a35520dd2e747b26e85b1ae528","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"9eb19ee24b508e691ee426e5ac1c4390c0ac05e3a5f2af426461294b43facdc1","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"2811bde8272d63e4a010b68f8d14d3f39481aca4a4b71e4f47de77bf7989e421","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"452cb7ce67a16b2467116936fd17f50ba519209c1f9cdbd5c33bfd65cb367178","bytes":5111}},"requestSha256":"d5887d0a03ad2bc41aae41ccd6479beac83684dc21a5737324c4104138241b72","metricsSha256":"efbc43a6e52c27af800592410bf32a60077f44bf6e24029f301d1cce0e6c1a31","continuationSha256":"f816aa39db831345d0092fcb20493df260465ea1c4425bc05aaa2c4e445e0199","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"2811bde8272d63e4a010b68f8d14d3f39481aca4a4b71e4f47de77bf7989e421","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"3cdf0e199b9134e47cc91947ce6f61597b7ae75d43629b33fafa96251f1996bb","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"244052dab9201ece5211209c2bbe18700cae4838a91b091e4a1b313c2d51f79b","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3b318251c23eac7a41e111b12a20e722cb594442edaeeaa3d62659832b66a6b0","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2459000","recipe_identity":"78c12c9ad6509d3e0c3222bdb4ffde2f9b95c911ce8a671a5d7bdc4614cc3d4d","loss":3.408059286757931,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"c2439f7a1d8e2e67750ca955717cb89a338a83d13d19971dd794ecce6ca3ad35","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"fe37ca95c51356a0cfae020943c96aacae10cd811a1176457821f81ce1fa82ab","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"bef254e383672305ce13590c12d80ba3ecce6febc03cd060952e24310f040e42","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"f0e8d26ee6dbba86c5b5368e11eeef7fa29621d5caab1c3e16f959e9028501c0","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"7c600467a29b628fb160fb781fe15c64a01bac7742fe964e9b92a95a89e3aef8","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"472ef00abb5714997cd5915a95a5154774d0f210ef7ae64c7457e7cd5755c446","bytes":5170}},"requestSha256":"8207c48749bc0ffcaffd20bf9a86e22a953c43c7a197df81cf837ae2084b26be","metricsSha256":"4a2fd654a1325c60f3122022efefae3e14e82fbe754f0083d836ccd3ecab16bf","continuationSha256":"7e56ff685ce5592bb27d8bce7c2952de6c07a8ad37339c5d737164e933c27adc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"7c600467a29b628fb160fb781fe15c64a01bac7742fe964e9b92a95a89e3aef8","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"eb14eea575cc6f4dbb735ca520fe85cb75393ad1f6ec21b4ce6231dafa599e3b","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1c052013a580b35d21b6f4c74b6206d9b04e43be7c80f95448476aca1eb5ff0a","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"78c12c9ad6509d3e0c3222bdb4ffde2f9b95c911ce8a671a5d7bdc4614cc3d4d","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2459001","recipe_identity":"2509b5ea32ef0ffb0ebf005ceecc156f22cbe6d8f76d25c10686b3a0b244096f","loss":3.3827230932190484,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"9e1a44dd42099f558045f5f7b40780f9868278d1fb2a4b57bd9db7d5d9c8bd10","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"b4a174bce62ffe762571d3ae5a8298af538af424ac86c7e21774c2e28812fa28","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"4e96d6ee1bddd037fb44e385391b2ea38df2323ec1b41af2aaf5fcbaa6b25a54","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"b62f4eee67196568deda65f914b2e3402da6fa8c843dafa86be5f7d3d971e591","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"974a85727053363e417abe173b0b6d52f0aa6334f6592afe976dba189b4a519c","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"6a660b0ee814f09a0299ddf114491785bc1e9423696d519649b9ad87575b51c1","bytes":5199}},"requestSha256":"1a5436928e7fa187808328026bb52bfa95d0f27b8372b17db2c08e6aaab52647","metricsSha256":"f462b205079d3252e934921f0efabd124b8e7b5b8c98380dbaff88172f13459d","continuationSha256":"545989c8a0f688898545ee0002be7602de237b4366b4a9837751cc8aa2432e2e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"974a85727053363e417abe173b0b6d52f0aa6334f6592afe976dba189b4a519c","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"4ba66f6232462a086bf49b2de5915e8fca58c92e4f700a698701970fcaa71f82","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"2f6405e8f779bfad9dc3b2579c9bc7231f7546e066268c8cfe50e51c41a41344","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"2509b5ea32ef0ffb0ebf005ceecc156f22cbe6d8f76d25c10686b3a0b244096f","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2459002","recipe_identity":"6f5a699d87fd16d4dd161ae06dc8e50b1f139a4076e30be07feb3f23d548ee10","loss":3.3892127433880943,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"579830805e4c9084350a6793e6bf53c89f6065f263dc7fa40f495c24bb3ac451","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"a11af1e3a42e4d91dc5a9be99321d6bb4cb64bfd640413c38d25b3e7eaae05e7","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"6b1588ca72f5723b63dd357a5769adf05780f30c73650f267ef85532380ed0ce","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"8cc341da261216e150773bd916603b5fad2c08319cfe8bd46d424c9ae4f664b0","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"a835f56bc850ea7b553c5469a918a5194950260dbc073c8fd02505fd34708cd8","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"2cd6b14934cf2b0b2c5644bc21e25b8485d8b024a5c817ea0aa1e03e703d26ed","bytes":5199}},"requestSha256":"a7f416f7d0ac1c5ef0b856357636ab24fcc27f03860186127d1dc85ac9206d63","metricsSha256":"16992776700b9dddd4eb1433d5d15d89a05113f5fe99f056ce89070a3a71d84a","continuationSha256":"07be180eb97ae42cd9a8d7e5ede52212f3f2a25e4566dfc875f00f8fa92e8141","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"a835f56bc850ea7b553c5469a918a5194950260dbc073c8fd02505fd34708cd8","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"448035faafe4144b6c4a1bca7b6b0377ace84fdddf69b633b84161de52fa1817","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8b8e9fcc0dcf358804bcfc043a21c1f43474cdb9c368077503826d23008ad177","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6f5a699d87fd16d4dd161ae06dc8e50b1f139a4076e30be07feb3f23d548ee10","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2459003","recipe_identity":"2fa4ab3b5a62966b8d21fa0e18a544e3eabdb3e2b29ebdbc2c7c69964acd2fc3","loss":3.3905290388975198,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"81e13027665b9cdd6b6dadaeea2de6255e0aea5a0dbfc0d851d708fd6bd577e4","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"a0d48fa6756c0290dc66192a0bc2914038f40d3dcb13285dfe14b313ea1beff1","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"c3c0547a62baaa559f6f550d4b39cc51c338d1a4c068b45feec3d8b2a668c64e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"df9f9955767b038898318b74f1e33a941920ced5249640c44f21a8613611f024","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"2854291f5978b27456d525a2ed81659ee8c3119141eae0183db0a824516d9e1a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"2f036afe3015f469261732a49cc19a361886f1254c59e068f291ba3aedb1825b","bytes":5171}},"requestSha256":"80bec613e2c60a79aa728db0a2cb960cd5461ce00cce436a5a54823f7016dcdd","metricsSha256":"7374fc14104f2407ac825dbfbd2dee4b13d2118c74850ee45b0d837d32dad04f","continuationSha256":"a01b2a7e6d3edb63877267d1366e17f7f59089e562e3a0dc5646d443edbd666a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"2854291f5978b27456d525a2ed81659ee8c3119141eae0183db0a824516d9e1a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"af3dd8f194a2ff2751e5711d1367b266bfd58d682b1efe88c38629976b6358e3","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"97c6639848cb4eb4b38489a677a87b5535bfdcf6f4ebd56f824f3d6d39efeed3","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"2fa4ab3b5a62966b8d21fa0e18a544e3eabdb3e2b29ebdbc2c7c69964acd2fc3","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2459004","recipe_identity":"b90f9eedd876f34e9961dcdbbf1a06be1214867e3222bf82039facaec7d68cb7","loss":3.3958272631829294,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"7f61d40eabdbcb89b823f1b7d19f15f11c1013c6707b35d9d63e497f966c907e","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"5ad3610d5ef0bcea1a1bb5b5004b981469a9d9210b30242f4339eb450cf00123","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"975daf3bbf5e3e8c25b239e16a18176349e6233b27da460ce7d9ac349b517e4d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"bccb9b15d7a0980d79eeb9a5cda039d659abf28dcb2969c9f2c31d023ba4a632","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"2cbbf575fccc7af0dfe8dca370f92f80584863eed40b9be3420dd7a82710737f","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"21d0cd57c66dc2426ef8f5359599cb85c56eda3a9c66e52cab7659b000990c4d","bytes":5198}},"requestSha256":"0fe3c7ed01495b592b2240679e5a2c34edec38a8591387130f940e50c9de50ee","metricsSha256":"b19e83970508ea91d2c07110c8c8d52f7680273a3f18f9cfecad0ef137855f2b","continuationSha256":"5cce4a9f8b6d0ddf1973a75fe6b2b99f2ba5daab71e2c4cba947f9431c5538d1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"2cbbf575fccc7af0dfe8dca370f92f80584863eed40b9be3420dd7a82710737f","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"3259711475bc1fd1350946510460bedd2cec92845209116041304fc853b66c5b","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c7b387838d2b2d324bf3efb81059fde2d66709c61b4e7b495a211eb58866f6a6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b90f9eedd876f34e9961dcdbbf1a06be1214867e3222bf82039facaec7d68cb7","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2459005","recipe_identity":"b72ab8427c791c09952459ad6d404041f966d43c49a0976833474c8b17e6864d","loss":3.390542539821416,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"88e0d819ecfd016ee5db0ccacc9a0fded450be58944e4a82cc7bd1e522d2e009","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"ff1c5170defe557d8b69a25f3cbaef33a3639b83e2248fe33a1ad55bc2b9a7b5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"03a7b0d6079e10adddb97f4819e0c278b93ebc2f69eb48ed1fa0feaf18ff0e14","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"265b072b1089bf09cb26741f795c4e843fd312060079dd77b2309bf1a1485cac","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"80ff3af5369e59d8be8e150ecc00e75a234f1c2fdedf4dd31cff6a90c7ea24fe","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"b1a84c10ed2a1380c3843146a363b0fb95eac54fb1811dec665ae23513a0a7f1","bytes":5199}},"requestSha256":"df2d85ac8dae75f233127d6964173b614a4053d913d8c9a8b225c29d8930428b","metricsSha256":"9aa23e1c5e5991018ec920c46ef47ed2e05bbf51f7f8273989334c686aa88ee9","continuationSha256":"55317d837d4e2ec565bfeaf82912304e36279be123e21aceea6db28fc2b5dbfd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"80ff3af5369e59d8be8e150ecc00e75a234f1c2fdedf4dd31cff6a90c7ea24fe","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"2e272dfe253f8a0171d634fd82b591101a8c92c248e5992b1349faa77fdb632d","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8d5ebf9ff14db86a803c1b7294672584ea1b8d41cdd62811b5125b477f1dbc4d","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b72ab8427c791c09952459ad6d404041f966d43c49a0976833474c8b17e6864d","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2459006","recipe_identity":"ef9986b011c4f0c1564defb1e51b93652038550b3cd8cc3f15ff5dd6003dc94a","loss":3.3953619767770142,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"40718c97afc4ebcb262cd28a043d761b1e9cf5f4beca90a577fc9d2709674810","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"04c441b9f753f1945cc4723c6672e194914ef95c5ad686fdda3f5356ef2b1387","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"36b3443760d3d23768465d6cfc0ee3f21acf4a79fa2e0e4d658a5089c49d8436","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"5e7ce16cc9594669ea141fa3f2d9390f792ddeb35e427ac5a0fc540a5cbc56e9","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"5436f9df25670c61dfd9116bf460871d51c837f531b57ac85b2a04a63761c15f","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"215c34fa775013309033928840ffa270e905f92dfb2c8b6946a594ec292a20e3","bytes":5171}},"requestSha256":"ac9b5183ed6ac169eebc15eacbaef2ef6eaf25e3d3623931d63cd541a9f2ca52","metricsSha256":"3149c2c1d1cede52b84762b02f31d654297e95e10ea8bc054937069dfb98a783","continuationSha256":"a2acb01a6fcf1e2a876f8b344b6d3b2b42018c90a45e96d369befa271e393ec4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"5436f9df25670c61dfd9116bf460871d51c837f531b57ac85b2a04a63761c15f","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"13fcedb4fe790e17f033cd6205f8a537b80d2913441ca798d4d39eb26fc9141b","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3888794ae6a812a624d32d79cbe9b18a96408624bc711130e4ae7e4a7ec13370","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ef9986b011c4f0c1564defb1e51b93652038550b3cd8cc3f15ff5dd6003dc94a","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2459007","recipe_identity":"550dac5eeb931c2fe5f59da55409dcbead9999b24264ed86885ce5aaf34347a5","loss":3.3957562129668304,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"7ecea68e1a16e914ee6359e8fc18a0c90ebfdc01b00275c05b2c3d415fa39d0b","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"8cca120c4e4b8ad680738607217e520e152584f85564f0b6228caa6d4368af06","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"27dc408ac776f0af39624fb53aca20640a9a86543da727a2a5bcb8f193ab3068","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"15357c6b7f18eca6928a433540d26210706cd1fb2799fea095d75e9f840853ad","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"f12a4f8dfb57d40ebb207412090a955fbdd2d03bc5e63e7425d0f5ec9f4748d6","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"9d3cfd15780bceb0053744c0b510e2af2c1dfc150dacf8bba58a94d824dbc4b9","bytes":5199}},"requestSha256":"afa2ae6f1124b5d7edf4eeed35a7925ddf324639620d715d68cfa0c6dbd1c7e0","metricsSha256":"6f751e0b319755315055fc6ac9fea5097b87888f35b94d17225ce7d462c2e9dd","continuationSha256":"f93f653fd67440293184b5b4b03d3ed7e836e784a2bdd1000a72623d981c00a0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"f12a4f8dfb57d40ebb207412090a955fbdd2d03bc5e63e7425d0f5ec9f4748d6","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"061d75fda73982bf75630b31523256e72380b50eda2b05a7de1fb16b9781071d","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"418c4ef9f43f2bde0daa6ee28eee9cb6da3a015a126564844aea51931abf203b","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"550dac5eeb931c2fe5f59da55409dcbead9999b24264ed86885ce5aaf34347a5","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2459008","recipe_identity":"001422b547c59960507e6eec2039a76ca445fbda0f74f6dca72d8fceb17e6970","loss":3.3862477731450116,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"621279060b5efd630b8c103da0d484a3f979bdad4df38197be42990da8d68334","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"bfa8ee1d32c1182359249b9e052b711a859e97443e67cb5224b109edb8264edf","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"77b0910aadc6907ea1f780742c176e6bb4e67399cfc64c1faab021db75758ee3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"184cdb61f456a97c72290236f5e6404914d653227e1dc2eca2807e4c804ac55e","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"a2ec180b6ef552fa2696130a04cadec6d836df2308042fbc9bd7e9feebe4f677","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"35dc7a0b4f4b123eafb63f75167db47f0fa6b26986fcb570dbdf644010a174a0","bytes":5117}},"requestSha256":"afa082198602a85e3070ddbe6bdcdb992a9d803db50d5a45ee689503d4e88ab1","metricsSha256":"ebe11150b19a6a1299b82d4cf43f02c544b420d502d86dcb237901acab00d0e3","continuationSha256":"7d1dc414d47a3a28b92d6753b3ce9d1ee1b4533103c10374f6b3ccdd9f6b5668","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"a2ec180b6ef552fa2696130a04cadec6d836df2308042fbc9bd7e9feebe4f677","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"a47e492ae3865055c507b85da2799f9d0ada70266362f163a6eaabe8af66ddf9","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"2c1360c99b9dc7b32dcbe5a4e8ad4b703ff086f008fb45a44ce9425071c2d45f","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"001422b547c59960507e6eec2039a76ca445fbda0f74f6dca72d8fceb17e6970","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2459009","recipe_identity":"34bf5bbe8b405ad509a268bd545cb8d310c46b8919cc0e74ac24fe98791c57cc","loss":3.394172872671825,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"00c25db775f150d74660a4df33f5fb15374ab9052d6030c2a674f73c59059082","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"c09916e99ffe5fa3d7c5b8356525c19994dfaafc16c42488e3d82b6c8a730b30","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"a11a84f43a0ffe732360784978dfddf64818c271dc22253aa9e83ef7e9eb1a1f","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"2f37016fc9c81ae38e5a448d645ab0c8e2d39060451fedcd9001bcd1e04b5ff3","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"08f12ed051bd0c37e9563630f60bd3a625099fd4b14d51dd7a5e53cf49f4502c","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"77d13471e6e47b6b36225c101dcc1f3ed59e43348edd0290f374a525c68f87e4","bytes":5205}},"requestSha256":"9f156ac2a6d66e9d5cd1a9aca374b32c8896a1dcc86c5eebb2b0d5ca083064d9","metricsSha256":"9d1ec8e9e76faf66e13b47812d51a77187fd16c82b08d3d25438082887e15157","continuationSha256":"332893f4c0325aad9280ba9ae79c5049b3a5125961bdef0395b69a64532e7910","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"08f12ed051bd0c37e9563630f60bd3a625099fd4b14d51dd7a5e53cf49f4502c","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"b1ffdbb3eb0de50fb7e05b50dfe62c73289da7dec67d3df28897783087947e20","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6c46b6f57ecc24b4c5ddf4565637d0891e5948119b0b330576f011545ccf9da3","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"34bf5bbe8b405ad509a268bd545cb8d310c46b8919cc0e74ac24fe98791c57cc","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2459010","recipe_identity":"e2908dd8e95cc9990312290396d32a4fe53eb94bad4a66b88e5be32e165c50e3","loss":3.386918653844721,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"c713b72ff2cc791dbb45cedf60519bbc6ab415390374eed22cf9104af7f9edd9","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"775993c093a92731421ced8da04138c28d1d6cf9d7176274665c8c3a6f0be1da","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"fc44fdb1a03091e6079c5865897a89f609b4821f5474fb3d63d7fe24b1fd8458","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"787c6998c8c6a2d0abb55f315722c183b768b6d7711b7bce44ac50f18b7b084c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"c63f7fe04f3de7dd1edcd131bbc7d673010026410f38efc75c69ff85356c6c47","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"4e4bb9c1c90c2a19224f4c393ca8707be31f0764d5d3a0e3f141a23232c95fab","bytes":5119}},"requestSha256":"b116dc2029213f3bc2c9736afe3d35f828fb8b8cf4a032b95a2c45dce9700d28","metricsSha256":"6f9da35aa25f378d85a226cfaa63df8c7b1f08db1fe28dc5e6d73161be0ff474","continuationSha256":"33c60cf3591ffd7397a8f72ff8de25b8576e10142ce53d425af1dd9d80317dbc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"c63f7fe04f3de7dd1edcd131bbc7d673010026410f38efc75c69ff85356c6c47","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"4e7c1387bc3975c803ce36a8aec6d472e98b3167f4a3174b13efd17dd73709bd","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6b3197b0c883a8e9a85ecf7bf586da7fe161c9b1438367331756abf7789e9f99","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e2908dd8e95cc9990312290396d32a4fe53eb94bad4a66b88e5be32e165c50e3","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2459011","recipe_identity":"93f692d94b5ffa374c408994a92da0127dafa06e13ead94f3421b1725b5038a5","loss":3.399715728285496,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"3ec063f1804f874520d371c0abc1e58da075614974f741bd87f5f9032170f634","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"8d48dae8f994fbfd7e15193f12e0e764d1901b42c10f6970c26290da1c200541","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"19d008fe1ca8c3aa965e3879f243ee8dea656f8189082a295d14b6c1db01708c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"9c5e833e9b052aa0fbfd2c588f754d34b8fb011c79601e6aeac63d4135f3bc5d","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"cd5a3c3badc3c4c854f535422f1ad5f7fc3d984ccb98e5a9374c88910bd99e5b","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"5d83dfd9d8913065bbf53a1b944af5a436bf8ae6f4d32e03ce5627c365f516c5","bytes":5205}},"requestSha256":"371022e3d82a016164cb230fa4088dcef730e1d261773e161ec05fa556cfaa8a","metricsSha256":"1584292097758ca274ad8d05acc14df92b93ed85008cf7656428cf12b03b887d","continuationSha256":"9da27f7ce5d291a2486038205b2ca1179f7114b1f1621849db6bbea8e9f0ed7e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"cd5a3c3badc3c4c854f535422f1ad5f7fc3d984ccb98e5a9374c88910bd99e5b","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"951a6c160500508821af2626378cafbb3474a7db04b05be05b80f231fd2de3af","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0513547ebe96c2504ecf688bebccbcbae61fb1453e075f6d64b185ed0742e177","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"93f692d94b5ffa374c408994a92da0127dafa06e13ead94f3421b1725b5038a5","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2459012","recipe_identity":"cba9603ed965bf533deacd8f349ad1a6129c48bb12d53bcd88bbd15d14c1e7bb","loss":3.3833897058015197,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"3a5775e25f7830fe09050e3198fdaa09d121dad3187d74a1721c16080598d07c","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"8b29981a41ff03f4216af71c777026e69530e1b152ea2cca22f68a6ef9b55833","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"7012465a2ddc7a328d0dd2cb7a71040850e6d8116c98abc371bb6c1ce9e01640","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"87e3aeb10f31a05a1ad1d68f5441fc651ca998425034d91cc1fa8d473c027809","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"e90cf127e310efd496c17a37662146722277fa250c254a8af7efddae702c1965","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"e83966c5ab8389a6573a00ebb4c3e3bdb5932a575c9908624803501ead3f7a0b","bytes":5205}},"requestSha256":"492318c31b3d56b98bbb7e345822f7e5cf3baabc38ef374620fe729f3a22203b","metricsSha256":"4303dd169791d69ae5e7a3cfa3b9f4f16beb86d24f3f40cbcb73cae1ace6ba90","continuationSha256":"f765a459ce6726d4c63b1e4becc0a4fcd84b815303931338bfafb3577362c7a7","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"e90cf127e310efd496c17a37662146722277fa250c254a8af7efddae702c1965","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"bb7636ce6413cbddd5a1168d84299b9577a068a5d2a892ba61841adedcc245c9","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"defd4aa99cf6d2ac9c12da1efe0f74b5d85898ed15457b9c99c7aefc69273870","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"cba9603ed965bf533deacd8f349ad1a6129c48bb12d53bcd88bbd15d14c1e7bb","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2459013","recipe_identity":"8bc1f7d21bdc5ec83dd4becb872766371768dd18aba2ad4c577e5b51f633f3ae","loss":3.3853547655989265,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"86992fdd145d929af00fabe5a27b7d02f361e26bff32a6d63f33b89e3725d6c3","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"9ebadd19b19f01ee00370dc564338c5547d98b25e89aab07ab7184f32528c510","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"3783b9b842ad649ed93aac06de862b8aca5172ee7d7a963e6d3683b162e2ec68","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"10302acdd3fd783a8ea952a07f3e4939e9444729ed2fc96cefed4cbbf6b1ca47","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"7b6510671b51dede416f642c61d7c7685e3877d0017fef4f615a4b31a5b5e4de","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"4a5f0b2ecfeee9e0993d32e54f733856f1b8f79db272b5323988153d671fcef6","bytes":5117}},"requestSha256":"b22cd67b03ac674ebadeae30621915b5a0c273af01a8468a9ac4cc54d77a3497","metricsSha256":"1385ffbc6ac5071f1f4690e07717754fbf5d917e6763fc091bbf16d3f27dbf26","continuationSha256":"712e563b999df8e5ae8d9f2842e62447a1ae0b62404d72d00213a8a2c72555b5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"7b6510671b51dede416f642c61d7c7685e3877d0017fef4f615a4b31a5b5e4de","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"2d4746d2fc59750ac5d2c4fa9097178fc2548076848d49daef50097ec55887dd","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d2c8414f2fb82e9a7d1cfc92746d4582a5e8776ba107a6a7a5757db70c9fcf79","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8bc1f7d21bdc5ec83dd4becb872766371768dd18aba2ad4c577e5b51f633f3ae","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2459017","recipe_identity":"a479553590175c90429bd0455b4e23ab00c7e9f2a9f52469930c29fcf058fad2","loss":3.3898292523767624,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"643e20daa87ce8db2c26dc25cb3dedc8fd769af09ad7bb47e9eab148f99b8548","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"1b040d2874e70309383286273734f15d52b6680b363532f8f160c5858f3276d4","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"15d18e05f2f6f14f0d6371f2561183d7a06c4284f164d0ef7a4dfc374c4d7a4d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"442fe4c316b15517ca582f6914e6ef41f04b7c781d2503162c64b4e9fdb5962c","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"4d8b73aa47819a794c0b1e6ca30edf3068495f40c6a61d8822aeb9bc23b684c5","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"6bd0a2043a0f6e6b8c4e24f5e7607bb1878261d99fb581822b60838ddb6f8c70","bytes":5119}},"requestSha256":"57a25ffaee0b506ab328ad6447e82f5c154e183fe64cc22d1c442622da301f9a","metricsSha256":"9e810a6666dd8ba98125362db08d61ace4db9d2b2ae0fa65556bc29858339cea","continuationSha256":"482591ac2d9e9b2ad208ea2e668578b14865cbebc38722e0ed89250ca5bef529","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"4d8b73aa47819a794c0b1e6ca30edf3068495f40c6a61d8822aeb9bc23b684c5","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"04bd4e7f22ec0b62db349a184991e6c37ea73c0b83dc8db52c8837c7a1e81aa9","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f0e5394c389af0a7893f5826b180ff9a4ac9451394c4337712941a271008aa23","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"a479553590175c90429bd0455b4e23ab00c7e9f2a9f52469930c29fcf058fad2","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2459020","recipe_identity":"27980fc7e980591d2d83efb3e531a8f0c2adf69841bda00affa56d30d16cc488","loss":3.3894190289794257,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"23979e8a8f3d704636c008c03c76ffa63cb53ba7e423ab2a445f7ab8029971e5","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"7fd604ff9e6dc2ae3fc3072fbc7878f2d0b92fb32a5b162be8e82f5d6b2a10ad","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"cd1e2b6c4e6a004e307d09ef4af3d1e844225783852e62271b67d2a4dc3528b0","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"f5147e5b8ea3268f9d3f59d97f1b725b8aff5b6f121a96ce69b020df8a353a1a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"36c5b0a7d06ae5cad8ccee0060018bfed4796eb32ed04a317a057e9f7c5fd01f","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"d23c37d1e96679805ea5167ef9d5d5907a5e888559ae3a36a346ab1ec8f30a38","bytes":5205}},"requestSha256":"874ff4b624ac0bf0373f21b668a1616136c06d2c0bfdab0fe926e6111a32f4e8","metricsSha256":"d3956ccda75b89db60984c49a337d9039a1d3879b50b0b3d04083f34ec455c75","continuationSha256":"e4c210fffd9101c55bf7cc71cec22f7ccbd4c9460df3df8c077de8c9a73ae836","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"36c5b0a7d06ae5cad8ccee0060018bfed4796eb32ed04a317a057e9f7c5fd01f","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"c6d80623000210362180dfaea5848b5a4dd37fc7aebdbb5a8f7dcee42fff65e9","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e1ae2184f4151d958ddc58d860b413f994203b60c09e8464f7ea7ae7f26372ff","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"27980fc7e980591d2d83efb3e531a8f0c2adf69841bda00affa56d30d16cc488","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2459027","recipe_identity":"d8a303b6c8b5420ce0f369c110a74ec463a381fbc4d4e455efb904d9c3cede2f","loss":3.393733781347953,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"7219df2215992089ec48af5804caebf5026c25e2cc273b4df8c1ca7daaa1cbf6","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"dbfdeccf03464ee82e22fbe8ae635480196efbcceea06d52cfe7f69a972e7bfb","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"a2965e6c9d4d28532b0bee3cb919f7ace5012557d8eb546e9fb24df7836637e2","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"fdbcebcc626f0cb8bc560d20b930f2724823fc876c9b234d823dcf7ec41aebde","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"afbd78612056018d98798468bbf18ab649c1f59bc87cbd20a1bdd77065f80aa7","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"9f38100fef996585abadec81ed82a37c8bda4ebb10375090ca0915606db13de2","bytes":5119}},"requestSha256":"9619a582411f054f181a9aedc44a2b7436fb1e1d84b1bdb164a29443fa6cbdda","metricsSha256":"75785ef52fb56beaed043f37d4d674cc5fa0576376c9ffc56815cf9e28438ace","continuationSha256":"bd365ded725333e8861c226d556dcc5524d231df4d6a88894a2b87e4a0e88056","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"afbd78612056018d98798468bbf18ab649c1f59bc87cbd20a1bdd77065f80aa7","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"b356b58ab0e291e8d305802e5d1782d0b8944471b7c9a622dd005adbfa4697c9","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0c57936e6f97fb101036898590254a47bd259dad44b11ad31d242cf199dcaffa","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d8a303b6c8b5420ce0f369c110a74ec463a381fbc4d4e455efb904d9c3cede2f","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2459030","recipe_identity":"3b657e3115ac0f8e09be03cef695c271750ea0e401dc2b6f69ee254fba77d5c5","loss":3.389741085295071,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"c52f9eb2df3c85d365b771cacf57387ffdd5cb7ded1b16ef1956dba63f0c0d57","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"3ec9b28b00a26d44bf95fc2ba9c4de254a5dcd4285f6300cd53daa9afdeb192a","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"abe186200eff1ca049918064b4d5abb7806b423a91842179454af252bf6783a1","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"32a29005095b698807c09fe3d70be1b85e64c3935caa93456ea9dcc0385aafc0","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"df0a127af1092e5aa723f78b8c2693b1929e9ffe1525e9d082c44ee9996e2464","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"c5a1e042474ee6efe8550498c3f8f40e2f70db0ce67c984d8cd1d9c18aad762a","bytes":5176}},"requestSha256":"de4239977b0d2fb060fe8561ce339c7ee77bc49fa39955622cdf795a46b32724","metricsSha256":"92ae7ca95989dfd23bf8d06e105a4b719a436abac77eb6a6404a99eff206d10e","continuationSha256":"d4dd3a25ee806d25bd48c67b490d4e655285150788b6c4c551adb4fd7dd2ed7a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"df0a127af1092e5aa723f78b8c2693b1929e9ffe1525e9d082c44ee9996e2464","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"66065a5514332df81ed232bb4f5b8e9d9373cd6f3c71e170a93b67930d790574","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3adf90d7b05ed2015a29f06942899da4abbb5b2e574414a900e2deb0cd0c6b41","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3b657e3115ac0f8e09be03cef695c271750ea0e401dc2b6f69ee254fba77d5c5","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2459035","recipe_identity":"8d22563491db6a416885f837b83d89b37c97bfaf3250019121820fdc197fd480","loss":3.392415910235665,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"76b31a1f99d6f79502530ece887c68f51dc386d9af59e434f4702c43b754cd79","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"ae8aab2362c3146b461982d3d241f89344a3ffc3dab0587f2e17542c71176c3a","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"d08403db4775e8cbccb09e7cbb2b3ae30876756d298060c3572ad8700b2dd2bc","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"93023cefc6b34ca11322b5f364585b19a07b121162a4ff89f72fe692ac22b2d4","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"e5e001d5fcca90974519fe86d590261f9b2a5b446e952b85a448dbe747b0aee9","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"316ad5ce46ef63a1819792081f26b31ceff1908c2e4911f66ac9cf1ff8f18048","bytes":5205}},"requestSha256":"2094e279c2d11e5e6d22de512f0cb6156451adfaababd17ce746b2eb876109b3","metricsSha256":"bc67b3cb692fdc0b86d2aa431a731feafefa8442ae456b86a7b030d3f08485dc","continuationSha256":"0fdf7f9971d4201d888d5afc7a8348bd81a0e67f664bb4c678941104e7d59c03","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"e5e001d5fcca90974519fe86d590261f9b2a5b446e952b85a448dbe747b0aee9","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"1a0d0df6b184964a51eaba1abe5fb32eff17cca2d2834bb404d644e952acbc88","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5b08ed312d173a3a1101c0daf4c42b52217dbf1251ffee0dab557d4810cc09f9","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8d22563491db6a416885f837b83d89b37c97bfaf3250019121820fdc197fd480","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2459036","recipe_identity":"4057a4f77f29b4be0ae14bd3925242e1abae5c22463536e95355fb228a28afbf","loss":3.383647938076112,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"fd269c58af761a04f0348d825a71c7bd11b0ba1c62441a5be49f215daebccc28","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"48a1d61eac0bc1e046cb2402cc67f5b8341e0b29c0d71f19c1c25610091d8b6d","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"dab6a5c69fafdfd7772ae527d77eeb463af2000035d7ff9046aff185c1e41a43","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"df2e4f39e8ab48cdc3578abe25183bf7b41a2fce1c525491f68866260c672fd9","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"fbdeab103546901e3773b06b99a538bf355bf83f82e961f2bf53b1570bd81652","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"3a7a607ad904d3d386c2d194fc292349f312387b0dbf97c7611f8b8ae96bc0c0","bytes":5173}},"requestSha256":"065f19fa96a0f17b1c587da80165f9bb8cea1387aa69a68f015944207884affa","metricsSha256":"7a12edb3bcfd0e75d0426f43d07d039af25c60f869000b43d9094e5564801c27","continuationSha256":"6471a4f8274a026c03053346b6cd8b979bb46279e88405c769c5f54dc89110a3","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"fbdeab103546901e3773b06b99a538bf355bf83f82e961f2bf53b1570bd81652","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"731d3dfda6f89118d49cd76ec6e1c64afb346e569966785d731b4140265b2757","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"adf17e69c1a1c29050d4096fe1baa27941863bfecc98bd511bd758f66f339c3f","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"4057a4f77f29b4be0ae14bd3925242e1abae5c22463536e95355fb228a28afbf","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2459037","recipe_identity":"d41d73830b615b9b1501b4892779c6c95fde1eecfeac901e0e750af7ca95cc5d","loss":3.388870167473736,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"d90c01c6fb316a8c0dc186054987127c47e30108907a4b26aae66728a3e18d42","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"bc5b2834978b6d18285c1f91b576e1a2a363c581d43b3d9aefcaaa1eb2982224","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"ba0d11afc72a70a40ae86c70e53956898cb3adde038fab9ada027de09a1e5412","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"7673375b70374e3ee78952952e2e6716d3ead3eb40abc1a70ddf0e4da0bedc4d","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"8cac7fffe4805e413ba16a92d9d336186b27da7881485e55b325cb8b2f29fa2a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"54efb6d32e2f32fa9d0901165554117a65d9caba2dd0f94590a2025655573b4b","bytes":5202}},"requestSha256":"52b112852741c090ec04d990f1bbc3bb5b98146149d017296bd6d66ae28c2ad8","metricsSha256":"f0301c3b1ff5e7706601cd3a7f9cbcba4a5218921d4894ba0642af8806bce2ce","continuationSha256":"0fbc2a95a945d26d7200840efb85ec2f51e93ab8d443b26df5a75565c5964d19","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"8cac7fffe4805e413ba16a92d9d336186b27da7881485e55b325cb8b2f29fa2a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"e38e1837650df402860ecb0d558d2f5ce69e3c6cb8dc1aa91f4cceabc306c73e","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"645937ef2b90ed2f75dfb2df714cfac0b9740dbdcc58d708f40d1a1aa03f26a7","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d41d73830b615b9b1501b4892779c6c95fde1eecfeac901e0e750af7ca95cc5d","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2459038","recipe_identity":"14d6383e74b9904ff2e2906bc6084d61dc053dc86edf2b92c931f106be0a95f8","loss":3.389124034119998,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"428e596bd32a03d38c537bc52d2d3485eb1de27dc0d6c05c3a2b4be060d5ac4c","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"14a8441e30c969123b37117dbf5ce1e3158da0a9f379915435323f098c182620","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"179944da234df6657010f55b881906f5403d71c132c493ed44aaf332f0b1c7ec","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"fb6f05e03595951ab000574716a4b0dfec94e3c7db124966289d918863fd2009","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"4636c936a97ccdbbaf830b9239a2ba1a04a3bd0b45d5d79bb51b774de62af6c1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"23c8ed9ffa463a598022495f31c772672526ac2e876fcf191ac49d7f039646cb","bytes":5118}},"requestSha256":"13730f1cf87421d00d45e2ca50919ede094c141375516ad84bd46273b9a73f8e","metricsSha256":"f361fa17941b93cc4ebbe8f5e9b992189fd45833107fcc37d9027b821ff9993e","continuationSha256":"0eb1300ade4fff324dad99c1fb0fa63da2f207e380896e99f53d55ac928b82f5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"4636c936a97ccdbbaf830b9239a2ba1a04a3bd0b45d5d79bb51b774de62af6c1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"3ec21ba9994d4ae771199283f9b5e451695a70ce4440252aa59844f338692318","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5ca8d1ca93ed3e3da00593aeff0086482f03c5516eeb05d6ee8640cf06b07b02","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"14d6383e74b9904ff2e2906bc6084d61dc053dc86edf2b92c931f106be0a95f8","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2459039","recipe_identity":"5dfc9c58461fa095ae4664bbae626305214a7ce854b4e4378e8bd94c48e1a24f","loss":3.3818844564960417,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"a2c5f160f9a40498137421783005c85a8f4d32e13646316b3f790567f0d3c6ee","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"5580f51d5a7bd20e7e8fed17829f45f8e6dc3539f7f4fa7ccb5511b390efa1e4","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"54fe74c56899f7b40f9eaae41e1d996e8e58855c1598a2c6ec628ddbf63b97f5","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"48ff4204b0f1cf2829f2b61c23fe28b8239f4575561c0fa82b33a9b0a28a846e","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"a2dc9528e6fa3713ff8ff709159f17ca086f4af5546958f30661b5585ea9d756","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"680a472260f37fee647a6c7921033b9e2e2b090d93d56aca191890140ab35a7f","bytes":5175}},"requestSha256":"3409f18533d1a05f19d57549e9b784aff28e6f6ad6a93d68ff8bf98389a70261","metricsSha256":"53567f7735285f59cf5cbd3b42b3d282d6d0e3ca63f5bbfc98914688cedbea4e","continuationSha256":"d68bf5bfce25bff9d83aa74e913ce493b3ed429fb2b4c81033bd7c50b853d834","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"a2dc9528e6fa3713ff8ff709159f17ca086f4af5546958f30661b5585ea9d756","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"67c718ef305980f0e81e1cceed2f1ac0a9b15211b4d0e125821e123ae17cf1c3","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1e3b203729b323ce163b556cc93b9bf71859766b6d473f576585d51757e52de0","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5dfc9c58461fa095ae4664bbae626305214a7ce854b4e4378e8bd94c48e1a24f","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2459040","recipe_identity":"29ca2b93585cea1c31a0d677cc69cd895985f21be578b47cfa5b4fa05440f109","loss":3.387048817222482,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"bd409a97d4ec7643b636ab10aea731a05bea18e3fa74a93f342101335c766207","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"2becfa5fa7f1f9519070153b125c01ecc533e5ef0633e947f589132d7aeb1e33","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"8aba8e98ccc18114003b95dd34c09e6a3c3e4d80005f9bc5df56a18c26fbf3d5","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"33dffed2c89a5b68a478a47bb997a46d81771b4a7d83dbf668d720275c04fef0","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e2a7de52f21aa0ac82f1342c47187eed5886130bfc4b076199a8fba736b806a6","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"05930f88c8df56a8eece1426d08d7aac2fb16136353bf69959e495d5bceb3264","bytes":5116}},"requestSha256":"4155875420329caa2ceacc0ab860f4ca1e7d04c4c4fd1fa9b2ec1d726c25636e","metricsSha256":"32ad4ba1e4c435e516096feb17fc84e0f435cf2aa126affbddf05e473b02b7d4","continuationSha256":"1d418b48f54718899ecfb26160630b330688fe95b3fd9237ccba16487d885464","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e2a7de52f21aa0ac82f1342c47187eed5886130bfc4b076199a8fba736b806a6","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"1fcb1b7d0017945a6049a1f267d760b30fd35f61ad0d566dff7ba27d04e0118a","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1efd5fd8b5b8265ed3d73eddbe407e93e07b68afd944e9742c1faf29274944b9","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"29ca2b93585cea1c31a0d677cc69cd895985f21be578b47cfa5b4fa05440f109","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2459059","recipe_identity":"1e32e7011609320f54750282acc761e329c9f33c3d20baf200a297c8202c8407","loss":3.389963399789087,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"207376c2f05f8cc38755ec416b7f10d0881dc26de078fb8bf00b6c26050c4f32","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"14f8d77198b4f4c0e79d830c5aa815fe62bc37488bc801224a96fc0d7c11fcb9","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"8bfea44b2d423428507943ff894280cb7964f384b32eaf8ccda4191a9cc0f447","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"c9123ae875fc791e633e2ea86ee97d97f4fceee583cc381544dcdddec8bbd27f","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"3fbe59d932291a43868691aa80fb082713e1bee1817b132d8f363e74d016cd41","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"c1180978efaa44302ec608a3b7d908d036275bba84a1594fe5f0bc44160451c5","bytes":5176}},"requestSha256":"8e3102faf827a67237faa6d38817b65b105cf6c049aa40ef03c9b179990eddf0","metricsSha256":"f27b2eb98453b9ab645904ca123176af9a637d4ab332e88e3e4d11b80c745a10","continuationSha256":"12f5b78956e986b23f85063cf2d36f1ae3c4e197e816cd6b88b9e18cdc3aa6d4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"3fbe59d932291a43868691aa80fb082713e1bee1817b132d8f363e74d016cd41","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"c6db604f35fbbcb8f750fe9c2bfce4391b3f2641e1501b7e07f474f2df56a324","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6aa7d90527dd17afdb748b62d248cac692496cc0c06172f4b9eac09cf5a03d20","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1e32e7011609320f54750282acc761e329c9f33c3d20baf200a297c8202c8407","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2459060","recipe_identity":"e3bf5dc98f3c46c45ac05f37d34fadd30149aac5e05ffaac7af7875905aca867","loss":3.3841094680773423,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"48a4e42b42ecdce8b49432222cd1ca0da5e2b411b55da161241e8fd13c577159","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"4c647cd64a9fe2193efd7b9f810b32289bc360a00c2db8fe0da78e10333eda6c","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"7db85f386d6f50db06f1ada60a2c24766cb8c838034e4b05244736629ae613fd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c611e5391bbe49effe5a03de05bc5165694e91a4f033ec39bd0ff2171fd0520c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"93356c3aac1c40167a336d4bba0099c4eff989fa68ef744e94857edc09c9163c","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"ae8f9071391fbfeefddf74ebab6beae0ec3b1b3638560bfa3a1da69f7f567be0","bytes":5202}},"requestSha256":"d62ccf87b59e2f403f12d6bf75c87f898c6fadfcd2230a540d63ff296de4e453","metricsSha256":"ff18515e06f9f5a5b2f62e7fabf5dbdbd408d87d91d483add19ec0b8bea7fdec","continuationSha256":"268cf28f5745faaa4880c58f0ee41250113dd32b0ee7830e43eb0bbe9579528f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"93356c3aac1c40167a336d4bba0099c4eff989fa68ef744e94857edc09c9163c","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"a1cd54194d5c519876c7516aec1376936332caa24384f972cd33357510afb0b9","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24ddf086f12daaa72f6c8f5c97ea6dfe84a66a4da3c93b78b6dcc330d0097bde","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e3bf5dc98f3c46c45ac05f37d34fadd30149aac5e05ffaac7af7875905aca867","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2459061","recipe_identity":"6d7daa3f0add95ada140751476fbb446152dbe780ce74d3a968cd6ba89413c6e","loss":3.3883398592945184,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"20017eeea835fadf206595eed18490ac1b4b2f80d3c37a6c8c78c6dc7bd8fe06","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"9c0d8f86028cf8c8405bdfb591e5e2b59977089f977dc6d4483690dc2002cc55","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"3c2aba6f771b00a9f1948659c0e3d579efdf79f59150129c615376a74265b4e2","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"e9931246718e06187eb366bd60e5b5257a1671d9b9b98cd9bf939bd24cf27ea7","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"b11c642f823e5e66258027ed0331f7916c0088f5d5a7d2958ac283ca1f11cf06","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"8766c9050f9c65b3973b6303f6434b72bb82f8de8283473ec02803c9318ef809","bytes":5119}},"requestSha256":"f317b537c3bdb5efbc7515ef3152a3530ae5ef5b8ba86253e225eb72071ae555","metricsSha256":"94f4d8b969b5f940e58e30ef33c683ada60a124d221e6a734cae825768aabe5e","continuationSha256":"a7499fc15583d22990b469af5e511df60547a014e8c7d6a884b9ae7d296ccaea","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"b11c642f823e5e66258027ed0331f7916c0088f5d5a7d2958ac283ca1f11cf06","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"a32c1f34bf404efdf47938be5d6b7e25890c8448a126783d7e31906a2e06d38f","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"adc23815adf8e2b13a725e575b8a7056b28165e2d61b4cdd7254f364fcc4a55d","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6d7daa3f0add95ada140751476fbb446152dbe780ce74d3a968cd6ba89413c6e","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2459062","recipe_identity":"57f548fad9b2b8a1fbcf0ee712aecbcdfbf49f5c8a9e8258ba84be7807dda428","loss":3.3902548419548144,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"1075080556768ae5c141a1adc9b6b712d56f07dc5fc85de3bab42cdd55949283","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"9335833c9995168ff22aed6e8cfaf9bd1819078c1f5459d4c533c2f5a50290e5","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"b48b2f41cad66fa1a96a10d44e3299a6da8f5f5a4dc2b7b5719d542ccd0adaf9","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"4f99d2f4a693684310cb7ea9117d85ddc7e21ed6830f4a50a7d12b32870a70b2","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"e9a65cdf3b8fb3a31f73c3f4eb06801650560ca5367ba5d39ea19b79db46cabd","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"41f08208b2911bb7b178fc49aed0eeeda7a4c24050aea2ac9ac566f39a8b7da7","bytes":5176}},"requestSha256":"6e30e7eb63b72d205fb906a357ff8bc4b58bee0e8cf00965808e05ec12d014fd","metricsSha256":"52f2929145b2b659d6f91e545785bfa2485e21f4824039443fe257b308519a58","continuationSha256":"a0d9b39ca9403bfcb321fb02f6c364b715d48dd2ce9868d88e1862b54db67680","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"e9a65cdf3b8fb3a31f73c3f4eb06801650560ca5367ba5d39ea19b79db46cabd","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"741d11ea12ff77c50fd4cc0dd2db2cc2b4499c73a5c671bf320bd796628c331e","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1d033d95cb950b45c886b6d270b7a529af33b21475ca791ad575dca7418be2a0","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"57f548fad9b2b8a1fbcf0ee712aecbcdfbf49f5c8a9e8258ba84be7807dda428","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2459063","recipe_identity":"419346579e38e853c98fff96f257f549653fe4f98ad3d1d7e3dc1cbdc5dafb3c","loss":3.3759750879053594,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"114e0e9caefa85b4aad4a342f26a82849b1375354b6b7467b4e2b0400f531973","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"4c4163f26220c1f6ba775236d6d6c4c4243e4286bf9df671dac626ab9d6aaddf","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"7b6f8e84acb03b1c1beab73a39d8ab9c5af537fb13e8cd31469cec77178e1fc0","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"861dae5027fbaa229a1d7140d58c4afc2a32abc10c5f8b2f40f62bcb229a6159","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"840d72d379b59ee0bed240a449dbd453c841b47d28bcfe4d7f04a258296bd831","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"712d4d81baa9230143b86f9c3efbde732e7b826240c50a977b3142c84f04fc17","bytes":5205}},"requestSha256":"59b90f1916f083e261cf077f9e9157cbe6c4680a5419b2f10833331082b7730c","metricsSha256":"3dac54ac94435f752547297677e6b482204ee0d79c849e7e75e9f5fd4ffcb42f","continuationSha256":"81ebc096d6f7d868512230855403b7b4cc1071b670950d3c4444343f446a227a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"840d72d379b59ee0bed240a449dbd453c841b47d28bcfe4d7f04a258296bd831","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"6ab6297d2b10862b0e406da7de9db5f9639ca8286ed72c629d074cbc7e051576","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9167733dbe45d80547253cc54cd9a7b1ba4f0facdf3f3727484d33a326c40d40","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"419346579e38e853c98fff96f257f549653fe4f98ad3d1d7e3dc1cbdc5dafb3c","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2459064","recipe_identity":"aa8d78c1c0b4ec723668394ce49d028710b53c3d28e6380cad9dd82a73fd3691","loss":3.381783574629176,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"34efbd37d3faa3accdb4d1ecc8c748bdf09f4a9a078089de28edfa2a42e4be5d","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"411f325ac59e156c2cf8d99421057f93d9f21d1faf18bea6f3818b1c2bbd1c88","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"627c95d090ceb03451e9c00335cd5e5570a4a845c0e9ef62f6f3f392529fe9dc","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"4f56bf9b9ab2fc167060d25d7cb2986af4321272d6ac059d8e1ef50575aec632","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"39f7ce3164f7cb98dd0aa9d5298672ce890b35805a47f3250cd7f2df77e8e382","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"c02cf66d9909765e4def3e47ba8798bb2f40666c6e711b51ebb60bd277a93d7d","bytes":5205}},"requestSha256":"9a20fff0d6f491bfb49577eaf98e7a5a231866bd15f0fc149a14f0db804e4600","metricsSha256":"a97e9edb3cbabc2adb5964f2a3b12db22c7b2fa7ecba5e422642ed34e0eace09","continuationSha256":"0b8d0d77bd3229ce49c277e556d6c38b229f901866e82b65f9ec1e9773ae68f4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"39f7ce3164f7cb98dd0aa9d5298672ce890b35805a47f3250cd7f2df77e8e382","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"cf3747014e380172811c72304cfa522d8b1efd13327eb66ded64d8ed24ef68a8","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c63c8ab822ca0bb7bec296eec0626fe054965c5940b6d9c3e2608c18b3431017","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"aa8d78c1c0b4ec723668394ce49d028710b53c3d28e6380cad9dd82a73fd3691","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2459065","recipe_identity":"d58af4b92c28fe96189d6924e0634e7e69d9662dcb06cedc73651f8a99ec07bd","loss":3.380806040847772,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"d4437848f1a34c495cc976456f01246c85dcc5ccb1341ac39bd90b03fad01e11","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"4a75f48529b832f2fc28f653f702c6b84ebde950bdc588e0efde64aca1c8d0b0","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"0f0dffe2e6f74374af47b2ce2855981d48a256fff40af512bba6b970abcff3ed","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"5b09fd8dd04c8e79cc43d62e7d75927aa2631e7a7b7b39ca5dc23ec11ea2707a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"802ab37dff09f54673d4713e41d627ad057bc7a53abd86ab15987eed9340fe8a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"c5f1a877022623925184d828a32aded2ef1d8e07ae7cb28325058bffefe71ecd","bytes":5205}},"requestSha256":"4670a0cb053d5df2537b047b6e546a677fa7164501f8f32e0030e084708f6d24","metricsSha256":"7eca1e0f163a7127ef51fe08328c9ac3f1fd241b260cf13d13a48ae92a800bb6","continuationSha256":"1e52b9a5b38b5c8aec94139c17ec9af8d873c309d28b787bab55c8d5e0474df1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"802ab37dff09f54673d4713e41d627ad057bc7a53abd86ab15987eed9340fe8a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"c480f0539f1452463e291bc3cc2e6b6203031d58dc3bcb872d9c9e1bc6afd993","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8890d7ce98f1d566449596664442e1ff8e67813bb63dc1c358c6295fe180d24d","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d58af4b92c28fe96189d6924e0634e7e69d9662dcb06cedc73651f8a99ec07bd","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2459066","recipe_identity":"818447c2f231b63d5c0473a2d8bf1492d0a0dcebec675000de052a91d3ddf033","loss":3.375018283755546,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"348c77a4d109b2a264dd5d07d65c5e213c10ff0ff7ecf9407844ed062689b363","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"35625ca600f46c04e34a993917ccf93d57de596398f5ae57864fcc9dc3b591a8","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"4972a678f4e8c1fff1e320b0ad21ee78b15d590a9901e5e7a7e02228e7b6885c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"8900c0601131d0cb962874d25ead3bed522d8788f32703016e6a6f6efcacdf51","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"9a4c260d0c82cefdbb00ef08e53186e12194b48e6628653b2d3d25706681e20c","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"d9b9033caa2cc30807fa9e39372631f4875e8a9967ec27bdf78050d12d0a1d4d","bytes":5116}},"requestSha256":"18bb8dae19a1b2a68ede8140b1602cee77ceb065e81d0e0dee2d6e1840710fd3","metricsSha256":"267e4c58fe3141c2e9d92ab24faf8313c81700f564d90474c4f4ca254ed3cc4e","continuationSha256":"e780f8ac40be8c51c329cbb665c5cc6d45238a87ba1de4cad762155b7cfa0323","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"9a4c260d0c82cefdbb00ef08e53186e12194b48e6628653b2d3d25706681e20c","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"39d543fd4b7a30761f2909bbdcf71be7ac0a4c3d3a2fd6ded73913c17f84f4a6","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0d7722d5675bffbbbbbdf58303ce9a27b4c8e729eba2569ce8520d20a5119ecc","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"818447c2f231b63d5c0473a2d8bf1492d0a0dcebec675000de052a91d3ddf033","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2459067","recipe_identity":"996d6c7bdf53f9f65e8863e49fd7fb95772e9f3c14d02e5510089380c2122c50","loss":3.379192834151603,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"905afa9c7ef01bfc703ba975502c713787ae141fd97ac40597060b7197e2d433","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"2a190a6fd25bd7247dd35685759ea6b90defcd9ba4ba4a6a074b8bfe21ee1242","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"28fd1b9a65f56e47c8823e012caa7d0433b5175aea68432101bf4ffedee228ec","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"4eacfecd7d550f665ae941b957ea36b93c76edee593b413458fe951111fbd134","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"16a2c0c4cdb5a751378bfe2724398fa978b1a89065f85aefd8cf073db7e02393","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"5a9d0b50562433fae65def5ce012198f89bbd6e3525afffad97cae75070c841e","bytes":5202}},"requestSha256":"dede81b8ff910c606310e3f2e39606ddf41d8faf7b0c88f86649176569db3c49","metricsSha256":"8b9ae0113ac6177579c99c43644257997db05164e85ac25bf8cbe7abc59bc558","continuationSha256":"cd8d24fbbf24a0dcd4127dc022ad843029f840edff5277336a836289b7a11b19","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"16a2c0c4cdb5a751378bfe2724398fa978b1a89065f85aefd8cf073db7e02393","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"99d68546293f2ec35f146ee4617579bcef0364009ca4df89b1fc23d22ecbe53b","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b91d431e5dba0df6f9c35fcb7ce1bcff74510f55ea7379068ff3da70acc940f8","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"996d6c7bdf53f9f65e8863e49fd7fb95772e9f3c14d02e5510089380c2122c50","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2459068","recipe_identity":"17ec99c05e99be6edd20011c1b70454181f2c313f6fa39f9611c24a659373764","loss":3.380375006930562,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"08931d8206e9bb26314c1d9fba0d91303a3fa8ffea226f0fc3cc82ed25df1c81","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"ff71b628d650f5e3379fbe4a19141e21f41ac4d35e55a2b7e8e0a1ad787d9500","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"0d4fcc158777a98225d7d77184661863c3acf3274027169f9b78cbe98470ce6c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"6bf49cbf150b7f0030209f45d654d238b78fe484ae5d09e441d76f5c4c78567e","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"ba5fae78a26658a0759f1ac0dfba0a7ae31cc1438a4fe8539ac298bee627d10a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"ea5ce5ce2e5ff32c3dcb4d843967d2a4d2bae683332e9697286abcc5c90c40c4","bytes":5203}},"requestSha256":"6251fd5659783ca72007d4d49c3ee72d9ed5f16a53cbaa03bae7265a77a2c318","metricsSha256":"658ad47c7071d98d3d194d85ea08eb2d4546c3741b21b43a3fe63b5749fb9a2e","continuationSha256":"bbc630db5e35a05e11843468743780e9bc405cb5689df60ffd67d0cd53de21fe","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"ba5fae78a26658a0759f1ac0dfba0a7ae31cc1438a4fe8539ac298bee627d10a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"5485cfee47df3655d2b3b491b7fdba6ee2d32121186794bc32585454c4ed2f6b","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"51f143dfb4573d23b4e7809e94e281c98bd55d8f3083d4e2c4bc77d7af0755e6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"17ec99c05e99be6edd20011c1b70454181f2c313f6fa39f9611c24a659373764","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2459069","recipe_identity":"46fbf7a7cf74bad757dcf7a92e540de95a04c73d3bc141f8141c1e15488d6961","loss":3.3775112392312505,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"82faa1e007205d5db21996a9b203c3cce7316b3e9dfcc78e8ad0b7d536189b1c","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"7cace288973c95b77a88be93ae4a8fe1db100d623a4ed88916d4d82077e9fe02","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"08fc8cd73d82035913185ed447dbcba0c999afde645252e52a4948f119a90227","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"73ffbb2facc4ece9d2280ec628f5cf9bf6ac6e93ef4698e7fdaf4e9ebcdac56c","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"90bd0736d096dd3757fba19c3206bec340a01091eff396508755081388fb55ce","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"2cf996876451134866c5dcbdc5d0d1cb682dae5f1fa48538c2faab56c85331e5","bytes":5176}},"requestSha256":"ba03c4d348d0a4aba54994ee2cb4b490c1bb39b9e7ce2a3fc6c05d466b473e82","metricsSha256":"4be36e77522e86edf4880286900a7952afc75409aec84b90ca497260172b7f21","continuationSha256":"1bcd591c0ccba2c224be5a7187f58a679bc87fcddbf79df5e29d2635098fa38e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"90bd0736d096dd3757fba19c3206bec340a01091eff396508755081388fb55ce","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"b4fbc33606e47290a4c48fe7feabb787e5aa86fff0e3156a793b0d85f84c3489","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e49d6a19595dad1d4fc245b8fdbb0779d4e8718ad0bc3745e29cba5ed0c711c5","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"46fbf7a7cf74bad757dcf7a92e540de95a04c73d3bc141f8141c1e15488d6961","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2459070","recipe_identity":"11f5d721c9373ef666d9ef665b9f7e553eac1bb554271f3f4550e0642e1bdc5f","loss":3.387363717423831,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"482267d451af62e01778779c76fbbe9239b0be3645dd84635ba25a1c462c4df1","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"9d06946a4155a1ea9a6790b5eb4cb5e51a89fa6c478bcca5c9f77449b56cb98d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"b19129e131a0eab1192ead6ead8386f6fb1464bd3e45775086ea771365db3ceb","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"c369153198d4754c229ca334a5fdb25d1183e38ad6d20df21a4392134dad0654","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"3ef02c771e5a53609dc82d1e273912cbdf4ff23bd2f8413507e1282511480b17","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"86d8f197f5b8bc386af52d9f3ba90ae2bb5cb4160e21a6788dfc81c9e5898d09","bytes":5205}},"requestSha256":"200a7e0796f0aada1a7366ac6dac72ff3d8f8ac4a2a8b7a1690719853a235d3f","metricsSha256":"a45fe154897a2c253e3807fc1ee6695f26ea48005309709cdac89d3479a05187","continuationSha256":"1b91b73fc3c6c70ed40a5b5ed073b8da943759b33bd192c35b294ec84a8d441d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"3ef02c771e5a53609dc82d1e273912cbdf4ff23bd2f8413507e1282511480b17","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"ab86bb0338dae37a681abef7018e757538dc50acdefea4cb5cb082c4ab19364c","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"98ff93c4852447fae32a4bea110beddaabb15c63e0da884fbf217d8cf0547a9d","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"11f5d721c9373ef666d9ef665b9f7e553eac1bb554271f3f4550e0642e1bdc5f","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2459071","recipe_identity":"0ec87355cbe8f4d0d2c44abd0a376622ab52dd7ba8fb547e521ff86f97b02e99","loss":3.377901445556874,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"ba0f7ce188a003328337d85858402374d5af1ec48516161b9037f03f91dda74f","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"1b0a5f345afb40316b65005aa2f03c440a4eb14c0b6c9df83f4ffe9be6748c5e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"808f42668eccfc81e573675ed26d5fae6823f18cd241714643d25ce528bc8df4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"cd851a4cff6d42ced1545681535b5c5ffe583adf49124d929a471efe5f6d09b4","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"abda07640ed5985d26ee798f30d8396b50a198f28ace1e8696d4b75c962b1632","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"0c35d9190963a03fb03f94490b946727e8f345d810345c4d44e761b33e6dea68","bytes":5175}},"requestSha256":"396c934ed6710960721f97b033144a2208a82f02f237ef2cfa158cb7f368b975","metricsSha256":"d3b20cc308ef79f77a0f871870b91ad06e6621fabc313356fce5fb7a5b3332f6","continuationSha256":"c1316d0c73b5d16c209372e3fbe8b33e1a9a25e2cde793e57702a7bd71f0708e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"abda07640ed5985d26ee798f30d8396b50a198f28ace1e8696d4b75c962b1632","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"17698bad9a097f8731985a99fece474a15a4fa28b5588200bca898fb70233af4","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fc2e9d5bfe6c344df81d884bcdc8e7ebe0556cddcac7c18b2408ca1b0630167c","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0ec87355cbe8f4d0d2c44abd0a376622ab52dd7ba8fb547e521ff86f97b02e99","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2459072","recipe_identity":"35d9d14d81528cc0d4f3a0a5ab43367d58d8e87362e13d8dac66005ee741d96a","loss":3.3758311076744336,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"8ddfbd5a527514edd736d8fe562a148da299ef3077a004abcd69c8fdb6c1ffe0","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"24893520ea23cccb7185a5db38ea079a32d00f6231ebd83366289e92a5d9aecd","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"2c52659ea06877db5fc0d86d27a7b498c2e03d3f00405150072b895f3a62c56e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"f3a4a6970d3df2b2fbedd1111daaeb0b87723f617ca292baed4310906438da0f","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"81be75f860e5322f3278d77f8b2e621f69f1e06b15bef1f3bb7540c4abcc0d66","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"82326c9515590352601c5e804d992edd01b527dc77068c8dbacb4e44a3d0e982","bytes":5204}},"requestSha256":"6eddde7fde2300d7dafb6beaf6aff0d3506283d05e5a968a549f580fe78cc00c","metricsSha256":"f201be16c8b444164cd985be0ff323a27817f82ace72ffdd8063abd3a32cd408","continuationSha256":"68363cd6f979d64cb51b900cd9248c41a6bb5363422bb9902b7694a1d626f8bb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"81be75f860e5322f3278d77f8b2e621f69f1e06b15bef1f3bb7540c4abcc0d66","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"71f1b307bd0919dc5f8e41a79561121e6848ef2d6e2be472cccef004d5c68f32","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e6a22cfe38823c9e46d2095a18739e777ff36b0ec1ca27289197e07c069c9007","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"35d9d14d81528cc0d4f3a0a5ab43367d58d8e87362e13d8dac66005ee741d96a","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2459073","recipe_identity":"b9f377e12cd43d5257e4c6535b9d1ef3801f7967eb1e56884db48e8e86ac6578","loss":3.3782369164252612,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"5a1a47710eff43eaba780f719e28ebf85e6c0021735a928a6e21dcf8ad46be98","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"aff9b2855b5651927bd3eab0b030fe067b64832ced96a0574336ada2f71b8a85","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"d3c829977bc25f207797a743ab5e3e1c43f1485d844c4c2a8ac8b142493006c4","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"740de22464faa2337c85298f0278171ad7c4eb82535ecc936bb59bc49a1044fd","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"5aae3374f5323f0c66681cba87dd5447ede150183affa2cbe305d572e32afaa8","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"1430e54f843b288c5cb1a41b1907521a6b1081c2b6f2595985c4f7c0d3cb3291","bytes":5115}},"requestSha256":"cb5414716e864df5585cfcadf8c3c5c77be1c4bdff70c20f3bc5499611264375","metricsSha256":"a68d690c52f2695ef0f09c846483d80579865e03fc95c8f90c8f53ec077f096b","continuationSha256":"e810da8d003971a1cf09f6fd72904415eef48c627db07d60f4cb53d07aca24bc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"5aae3374f5323f0c66681cba87dd5447ede150183affa2cbe305d572e32afaa8","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"59bf16ad0408bf2861e5a5944904382bd981ff3d8f659ab0efcfad0451c22b12","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"29f9dd585bca19eb140978c692c59850b239137ed186a41202361921b4532c25","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b9f377e12cd43d5257e4c6535b9d1ef3801f7967eb1e56884db48e8e86ac6578","loader.blocks":1593984}}},"sync-wsd-h80--0.004--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2459074","recipe_identity":"d9dc9f344256d968910cdccf22793b28c24b4c527335beeaf0999474e0681b0d","loss":3.37813549995855,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"49e83c6159a8921728cabb1973b1da3ed3f22c2d605af0317bb164cf77924197","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"ace569eca037ad1840bfa6f75d57393d2faad5dd38c4096bc46182da2f2baaaa","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"3b7571d40a9a426729734669dcc758d56e46e88ab650fe33e3a64ecd2f2a4562","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"c55dd1ceb0192fe640cddc14659cd8976b3b807d703e178c011b2d662d561c70","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"51b632c5a090b4e2114e497f3d973a71c52866d14d080269981c3b7a200da87d","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"c2763971d2e8f521e70e38e5d8ad067ce9bb2b405534855578dc6589f87d9491","bytes":5172}},"requestSha256":"4c2ad775d10eb5c42073ce41c410376bf3f00536097f99b2437ae66ca03c45ab","metricsSha256":"df854dfb256d14c251eed52422304761722263d316086d639dbb43a1ecab9e1b","continuationSha256":"4fcc3047f75f77e557f0f153a9b02403158f2dea8e7c1adc260ef23a0bfb9ce9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"51b632c5a090b4e2114e497f3d973a71c52866d14d080269981c3b7a200da87d","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"f775948dad16d7e6e767286f02d6bb21239ed7d203a7c272dc9736449527d417","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"040a82ca6726124539b62558deb1a4303b18c050a7f42c4c96ea09ac768aed16","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d9dc9f344256d968910cdccf22793b28c24b4c527335beeaf0999474e0681b0d","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2460653","recipe_identity":"9fa14213ce387d491e6b37eab5bde48ded0bec1e48b4aa75b3b259a2e6e877f8","loss":3.371351940317979,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"18526d26a777e518ff46142503ebebb1cbb89fc5f9c80dedaf3478287bdde728","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"929dcd8f0c5bfdbe1c35d7da259e1638bf593c72ede8052b10d16502a52e11e3","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"b287d98ac1e0862836f293cee03b1b5cb32cb8ca7c20d2b88a919cbc8aa8912c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"ee0ce6846348d22804d239fedd2f04aa55b72785af48e8fe6efac42e1cba2318","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"65008a2d5bd1ebced6db7ca21b49ded7ea607e2119e81e19edc1566110a213cb","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"4f9a4cd662fa7712a88b62965b902201758b212b5de2790fa6d8d07e3f7c6e96","bytes":5117}},"requestSha256":"4281408c5175bdb84f8a60e80f2055aec26df9199753ae0cae87cf86b6b1d3e5","metricsSha256":"122614f24a030a787678c283ca819a0e3834e066b0af5498ecb61c05fb2ba4b3","continuationSha256":"2dcbb540d4e0544d99ff616d6c5884c9f017d5c906d6ec3eb428167ffcb14b9f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"65008a2d5bd1ebced6db7ca21b49ded7ea607e2119e81e19edc1566110a213cb","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"810038bb6fcd9dfd4022d087b3bea1ddf945bb412649f06798849059563edc26","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24cf9859fa78e12ea697974d6c43abd0570001d5926536c2e4da988c774c341a","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9fa14213ce387d491e6b37eab5bde48ded0bec1e48b4aa75b3b259a2e6e877f8","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2460654","recipe_identity":"082cdd3da7c4895e3529d6514715e4efa016c70241bd47bab2bf90298d8a98fd","loss":3.3773290357226062,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"eb8120717042710c9e6b20194940ee0e83cbb4bd471ba7208db9cf332bb77753","bytes":749},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"3e7cd8061f282aac0d8e25f84de4185048894ef646c3f31b0437b65fc90c8637","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"28dea8668ca572fcfcbab43c6cd04db483a1ddbe874ced49500437d45b0f09f5","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"e171486d4bfa4d6ca1e7bf693348d6d7a882603877a51a5c0fd21fdc7c18f61e","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e9a100e43e947c9b1042b7ed914769b7ff47ee90fc609fe27d00d77d77e834c1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"e0da0b272cd21f5ba8232363bc3afd873581cc46657d643650a434ba33493728","bytes":5119}},"requestSha256":"a99aa1e5f2d5e6cd27eebc9ad33681832f585367ffc04dd208389da9f763e277","metricsSha256":"89208b26466dc0fd10046d465ad22d5d8c26c73e1ab4741589988b313837ae2c","continuationSha256":"406fd6ddd10b7d5c3335c84f70cbf282296ff1bdc809e124344e2c94e79b1eb8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e9a100e43e947c9b1042b7ed914769b7ff47ee90fc609fe27d00d77d77e834c1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"dc411531be78f085fa1810be0d69180430283619e8feb5aafa3b6aae1cfbb880","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0584d7e3d6e23d96e4089284e7b144ed42e49ca3c66cf4b4ac21e73f79008ee0","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"082cdd3da7c4895e3529d6514715e4efa016c70241bd47bab2bf90298d8a98fd","loader.blocks":1593984}}},"sync-wsd-h80--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2460783","recipe_identity":"09b98b23bff6095db3084f2b7820643bf7eb063817e7948414c7cb57947891cb","loss":3.3756280799544163,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"0dc099acaf31555f3198f88d9c2dcb8b4839caf5153d67d0cc8d69d10072026a","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"39e61262f2f5f7ee46878d742f8f3441a472e93c5425b0cd62a073fe0993f11b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"fe1592886e36bc640b61742e1065111134030c3bd52940ecac6bd8147b656457","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"89c936a2ed865e943ae8a64c0c72f2c631fced0cb3d5b1ea135229e604d50ec3","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"9f7bc56bb57d9aef4d4c2d633df971837f7b514b8f4736506b7190f8cb77791a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"90d7fc894fde6dc1f5b805592a7eb073f6b93a43e9a7f2d48e235474327537b5","bytes":5119}},"requestSha256":"76b8783562078929576c3fffbbfc06ff93210d3fadd65f85abc4605f0a668b38","metricsSha256":"c07934c9455916b59ca5cc20cec6c332210622fd9632374b461e4065d4f567f4","continuationSha256":"4238beaf353968ba4cae9c58894e1535c8ccb6f15550d439de14e62a907e19a4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"9f7bc56bb57d9aef4d4c2d633df971837f7b514b8f4736506b7190f8cb77791a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"f3573e49a5a6269f95e0a241f9799a78050c14bb7d847a28baa05d2dbbb21fe2","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"23c73841b0d9e7d9869faf5dc75f6557a62959ff4161478fb981742d2573f48b","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"09b98b23bff6095db3084f2b7820643bf7eb063817e7948414c7cb57947891cb","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2460784","recipe_identity":"4acf9a5ac805fe1bc856d5c46fde88fd028d15646353cd7060f587432d3cf68b","loss":3.373592376293663,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"bdfe04cd65b9e55bed0fe609507d0670b75d7204e1832c308458f435911eeb1f","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"c903aa43fc7be7744fb446a08fa7b2c8e0bd0d70d9477c94c4d4382693cf9ca9","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"63249148738231a45df094ce89a67159d11874f7782cf0a01791f594ad41426c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"49878e8aa91cd34d5556eae2dba7daf54282b299b70fcb2af90a77ca1d7108ad","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"eceaa36844461e2478999b3f1436a3ea40a907b41faf9607bb765a3f575e55dc","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"1ce8b6e5f63df06ab30a9d3613bac19f3ecf35976f13fcdf8b6077085f7a6524","bytes":5176}},"requestSha256":"380ffe6861f6b5c22f055e1a112362e6a0a78ba19cad8e50dbacf8843bec2795","metricsSha256":"06e3d5853bc7028653f175c9d8a72b142a6ab02469862af5edad1369046c5213","continuationSha256":"58b27c78e0b96d76d632ef42d1f604bce7a3699be0121910e0635414ff38f070","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"eceaa36844461e2478999b3f1436a3ea40a907b41faf9607bb765a3f575e55dc","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"6ae9203d475a909fc04bc740a39b8ca77e721087e0623f1ff3a5416696a9ae47","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"663afa44c4a430a0efaf6c95b14587475e07dd6d71cf66248e917090cdc2db12","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"4acf9a5ac805fe1bc856d5c46fde88fd028d15646353cd7060f587432d3cf68b","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2460785","recipe_identity":"3ca6645273253308dc0974eab53881c87ef9c5d43d5966e7b1636fa0cd0cad3d","loss":3.3816051022697224,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"b6dcccd98df457cbf55676b439cf53f85af0b317d810cc437d098f73be779f26","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"5acc84a3a59d947e8cb5b59cc6cbe11787a5efea12fe14b3580541531810dfe6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"6586570b21c5c941d795a6255a200f7e834283692268bfd1d17fd3bb71da3bbb","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"3e1671d203a5ae8a0d57af1efdf8a36445cf3515a2a606f6c7b821b0cf649b61","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"220fa8683234fc1d63dd8a06b1dbf23a22514805f6f326cc40dcdeb78add38c9","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"af10a29d3bcc1680e6c994ce0ea5cb741856195a45b01ec35b8a9b610ee59d54","bytes":5205}},"requestSha256":"395acab9cc29f78369d51cfe996729843c44abbce270c85af8c67702f8f72fa6","metricsSha256":"9fe9634c4cd522fc56dab5f458541669fe290aa01de99809a88812a25bad18f5","continuationSha256":"3831b5d95ec49a5e99f573e7b5bc3ef1616fad598a353f72f33427b0d7fe0425","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"220fa8683234fc1d63dd8a06b1dbf23a22514805f6f326cc40dcdeb78add38c9","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"c48f25b3a75e58b05befc7c9d4ced2e4b9b6a8030a64ddc6fa67064af5162000","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ceecd345386516d1361ae215c344b1c839b14a0c465dea3931948ff59b0676dc","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3ca6645273253308dc0974eab53881c87ef9c5d43d5966e7b1636fa0cd0cad3d","loader.blocks":1593984}}},"sync-wsd-h80--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2460786","recipe_identity":"53a4d08086815092f0617bfc5e77dc956b4628bda4c50356d2405f5de4019e67","loss":3.3754881626222772,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"9adde3ebe28d041d8154e45a3a9978c99a48b8388719455de6e8f951a1491eac","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"78c430d6089385453d82433b4343ba44822fcaf286cc5399554bd64cbe63d01d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"708d214793ffd03b3a9ee6e150e21eee5b089a62a311dd8606620d0a5960c938","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"9cbfc88c46eeae5774c7adaef6e614cba3b57a9359e0fb961f1f8a14b0913a82","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"50fe7708535b4d926cc966939715f62abab8ab01391c6bc626977cc36b03e099","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"edaf49aa264338add1cddbff39d127a9bf22bbe3298097f4c405e38a0a0e59a6","bytes":5202}},"requestSha256":"2d1c2c80e58286fe5314e29f720f8f8730a14b489968b06e554f55e37f038885","metricsSha256":"d69cf0378eaf0d0782d668ea0405a3182ddd8ff63384c188745ed44f54f95f9c","continuationSha256":"a0e8b346bcb05443a96b0a522fd03ee70744538a1152e3eb743c34464eedc830","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"50fe7708535b4d926cc966939715f62abab8ab01391c6bc626977cc36b03e099","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"572407591a44fec40b859fc1f3cb58234ae4c85db5f8164479de305a03f995b2","bytes":244946027},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8fc3a65934a1ff2f23e2f2a3f23b262daa7d3faa5d8da474b746544dd2e186e8","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"53a4d08086815092f0617bfc5e77dc956b4628bda4c50356d2405f5de4019e67","loader.blocks":1593984}}},"sync-wsd-h120--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2461519","recipe_identity":"e98adfe88807b82f846d5985354d418a2109d4dd982a203100d8b0dc2eb2230a","loss":3.3691511967457104,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"eac7a3663b74a87b958fe6f238d8956cb21c75cdd111947db164964bf532e862","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"b92bae43c9eb9dde1b6b9c1285caedaf8907f767de19bb80ffb6b2fe65b78b56","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"9a1861fa5d0a98cbc74022ceae1c4aee236512bae1fd4eb0f7f8994b8a13c433","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"8fb11f02d9c8929f144328b902ff8d057f4677e4744c275bd7c18f417bd55829","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"9b3bb840968c1076e4845e07fc328ae7e8a4563f82c962754a0d6f3e59eb694f","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"c8161e59d88e2c8bffbd5427b2a1c81836048f0cfe1dae114bf963f9528e3561","bytes":5120}},"requestSha256":"c691d6e97cb6d8925746390ea5409c8ffc1e078cfb2d576483c01fd9afb7c086","metricsSha256":"4e9875c3274e16ab5be0e2bb596900b79bda254d34e36dedfeb1e34b881dd5a2","continuationSha256":"9d27ac2498baea8363a1165ec49f192697a9e23372be2eb73c074b4f8b93d61d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"9b3bb840968c1076e4845e07fc328ae7e8a4563f82c962754a0d6f3e59eb694f","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"d5134a8be1d8b08c52626e1999af4d49e679c3840bd1efb255716f4f836f212c","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1cd13860e821c339a3e07df7e607bc3ef38a7afdb1f67850da89dd516d72b81e","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e98adfe88807b82f846d5985354d418a2109d4dd982a203100d8b0dc2eb2230a","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2461523","recipe_identity":"723e3947164c972bd9a84c0544c97971b9457914078d31a9341cf91dba93203f","loss":3.364713668866867,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"2f2adf1e6ea7c2fea364f18d3fcad4368b0d2cd806fd34498bca7a9e35502aaf","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"b873e8ca95db343d149156d4e99a1d190548098c9c1efd36fbf5ca608bca079b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"4be4b5aed49f16a586611e0b959523c2c2c8ad92ad39cc8643d49cceb03e7d03","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d90dc2b9fc60aed35d1518b5a07fbfe6cb4bb9bbf089ba0c2c6eab62bd9440e6","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"f1d2ea6c73ff0363bd257eadca3912d97ee481cf0885ab16b06002cfd568ea8a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"f5fe1a3bb1db54d17da09f72b5dbdf7f24291555c59172da7158f4b6753beacf","bytes":5174}},"requestSha256":"69a7b46cae30b98b2581f745c99c46800dee89deb988f15595bdccece43a6e55","metricsSha256":"93280229bcc7573da51aa390eece57f00e835f81210b2b6e7d961b1458755b57","continuationSha256":"b77c5a0fc2e7217e90fbd266cfd6b15f314647086ec598b9d55c6531b4fd5b12","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"f1d2ea6c73ff0363bd257eadca3912d97ee481cf0885ab16b06002cfd568ea8a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"f428730712bdd22baef8782126a171656aa4c024da507c7f2af079fdddd860b2","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"43ce31887392f962466387fb450cfbde347d49c252662cfe1e74400ef3a115f5","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"723e3947164c972bd9a84c0544c97971b9457914078d31a9341cf91dba93203f","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2461524","recipe_identity":"f4ba144d1b59a693f5c7de8e60102bcc63839b7d88127850a54f7e370dcab36f","loss":3.368625202189785,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"e04a2e87778233c610b62fd2d6cc383f3de694e6dc700f87d0eb49c2c54c290e","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"1ae510a8f07b5d760ad7f4f4c31c39c9e81f066a410eb6e3bf8a563102625d2d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"60638de1902640ed3ece1df33fcf11bcdf251b83b19d4f2881f45821a50cdc5c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"3ded072f20d0203378db57fcece5b08e3642f5b55c3cc4c145d2255f3c402be1","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"52cb6e9eef4d49cd9b5a10f398aa0eac86ea3da33d0737c5706db380fb099da4","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"e61b8a98b43928cb27fedb9da0a1f9166e841c8178392cfee745eb89f03acfee","bytes":5119}},"requestSha256":"b278240bf47cf29f4bf25e92b78eb2a5878c08afa5dd51c31296024cf6d46300","metricsSha256":"b371ff5746bc6ada31e2c7ff85a382671b7434f057ea5e3568c13687f1c7decb","continuationSha256":"702047ba6f79b46a21c60d74bed8e55d98160b6100a67da923e9073569ab97d4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"52cb6e9eef4d49cd9b5a10f398aa0eac86ea3da33d0737c5706db380fb099da4","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"03249cf2d2950d9b6b3f6ae9a2b3743bf830fcbed92cac9b2a20922fc613defd","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"25f691cbff5df50ce00e22d786622d75a00855879a66c803c0e580ad37d3daa4","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f4ba144d1b59a693f5c7de8e60102bcc63839b7d88127850a54f7e370dcab36f","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2461526","recipe_identity":"a9486b29f35d32670ce9148ad946a15c630bbb859dbdce0dec20e61af611d513","loss":3.374923689294086,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"05dd730c9ce3fdad812842cffc43e1ab877fc0921d92f7de621cefd5e2a94d84","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"7428db138712591d217dc32a9aabddc8645ce9014c4a9ce465bf8a70a802a494","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"68b2ca0a48da57dc15414c5b0d1c8e28e9657890c40228a11198adead8718737","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"9ef80dba6ffe0002f5607cf8c7145dc08746b62f34ff4b7c68fa3ba8988ed41e","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"2528475b0d296edfd218d95c03166d6919e986871ce9fc972e187df75451e937","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"0d6d4661103dc46fd4ad8fe29914367e397d453ccf290eddc0102c191eca9c7e","bytes":5176}},"requestSha256":"901fefea5a3d3e456e62c6e75e7637051695c83e4722ef6a5dda0f4f24f2bed4","metricsSha256":"1b02a71239ff87c402f28feb7c7702d1efa1cb41245dd4f8189e0f9a7bf98415","continuationSha256":"8cce4ef7c3f03ea188ccb793b12490d62394ced632d3f9bcb9e128adcad84a2f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"2528475b0d296edfd218d95c03166d6919e986871ce9fc972e187df75451e937","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"2cbc2452b60206b714c22557b3f6b47b0c85f29c91ad329248b244d15af4d27d","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"c19be9d22819877d62c65ca64121218cd08372ce460640572eae2f527caa894b","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"a9486b29f35d32670ce9148ad946a15c630bbb859dbdce0dec20e61af611d513","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2461527","recipe_identity":"ca6a6cc701ffae059a2069539e5c0da265bf206e1d0a1381622541e6b1fc6f6e","loss":3.367432421913799,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"9c581d565a7b253e5abf05501fb59add088abac6a1d7d54ea88824eb349c379b","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"380971e8e856a6413d6d26021655d68bb28f33614f95624008a505b54a0c8771","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"9a76f0350df6a62961128375439ffa001d7e71adc47550a7dfb31571a1818c72","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"4c91c9765d04a5a5705da95b026a807bb2b10865948d4f632915e3f633dc8f77","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"9eb19ee24b508e691ee426e5ac1c4390c0ac05e3a5f2af426461294b43facdc1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"d3a59d4be8fa7370b9e37347851df1776add98311f6a2e01e11e644fde26eeda","bytes":5176}},"requestSha256":"f93af72cd26135e1abb8e85286bb6b31e4feaa1e8489926fa728bc344c25c77f","metricsSha256":"ff9226476bc96b1a863da239d5eea51a7f4e4ef1541798953c3712c444e4cafd","continuationSha256":"b5c4d2f223c3cf33c4cc7f65fc7e21fe401a677343b5cd3a00054c67abe76721","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"9eb19ee24b508e691ee426e5ac1c4390c0ac05e3a5f2af426461294b43facdc1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"5e0af60625f94bd7b0ab2f8cd0ad5312cc406b607b0427abe3bbcc875e6b3210","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3b318251c23eac7a41e111b12a20e722cb594442edaeeaa3d62659832b66a6b0","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ca6a6cc701ffae059a2069539e5c0da265bf206e1d0a1381622541e6b1fc6f6e","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2461528","recipe_identity":"268a9a4c6bdf736ca3979d01d5aad949dde85fe8f93efda5995bb980edb9d1e1","loss":3.3699094124197315,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"3ac132ce0a17c6567df338d06f89eaea4a7e1d3d8a2aa56bf99ade8b7fa5ca14","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"938d362f6521607d62a8c255b1245928e2f149c28cbacbf011046fca084811d6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"8007d5d246aa17b3a7521afb9902bbd65aaa5f6ea8927663142295d1a02f8990","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"010a7e064c5f6fd91f54dc55094a94b32ca5af7a4837f75f3bd59f43da28c82c","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"f0e8d26ee6dbba86c5b5368e11eeef7fa29621d5caab1c3e16f959e9028501c0","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"5451cda7956adc751044ea0aaa3d5fa1f86f26de4ce4eaa52542cc8744f4d948","bytes":5205}},"requestSha256":"53f21f99df82c724495f1d14a3e60b72c8732893a2d4e036ccb8c5ab6b21308d","metricsSha256":"4fd3236d6a37b8f92d19b52ebedf10b8650a7aa3a274ffbf3580a1c8498cdd1a","continuationSha256":"7749707357a1125494c6a687d09a238a14fe7490406b6d6b9dfb8de51eafacac","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"f0e8d26ee6dbba86c5b5368e11eeef7fa29621d5caab1c3e16f959e9028501c0","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"9606b0275df464fab6b2000060311e4b45d05924ccaa2071b8664fb349753c86","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"78c12c9ad6509d3e0c3222bdb4ffde2f9b95c911ce8a671a5d7bdc4614cc3d4d","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"268a9a4c6bdf736ca3979d01d5aad949dde85fe8f93efda5995bb980edb9d1e1","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2461529","recipe_identity":"fe161ab59b1da4324bf8a436c924204a0cf882f05f3305226efc7809a554ea61","loss":3.3589789181908407,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"77457fa53edf0dfbd725a486667d9122cae6217c3f3caa5e13bf1d39ddd14ac4","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"f227c9f5341f3847aefb41c31c04d2f4e4d1444963685e1b7c1c97e2ac4bee8b","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"406364fabdc517fd897fc4567125be094168a1da71d39467d4e311c3ef4b9f78","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"d36b8040241a797045387b016aaeb74a78909e9c149780c47b23a402d1a6e301","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"bccb9b15d7a0980d79eeb9a5cda039d659abf28dcb2969c9f2c31d023ba4a632","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"2be131b1a6b0c3aa5290121595dd83f4bbb8bddf49677c6b395bb8a286937504","bytes":5119}},"requestSha256":"9eeb4eb8b5ce13f441bc5fda3da5c3e866eb9907fda3297fe02c897a8baf16a1","metricsSha256":"4c7c9b80941277a6cbc5a848d0133f62d7b035faeb13b2e8f53954b6f0f3b008","continuationSha256":"5cf9a81e1fa6ed83fdc49e536fd8762ba0a724d9f26f7067983ff92354b14cdc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"bccb9b15d7a0980d79eeb9a5cda039d659abf28dcb2969c9f2c31d023ba4a632","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"c9e30dc8c5df4e1faaa469a2ff40e599a38c39ca3bba7caad441bba8a1fbf155","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b90f9eedd876f34e9961dcdbbf1a06be1214867e3222bf82039facaec7d68cb7","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fe161ab59b1da4324bf8a436c924204a0cf882f05f3305226efc7809a554ea61","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2461548","recipe_identity":"d83bd101187d23ee2c530a48d81c9e6473d07a4c3c48783a6129b33fb0b4337d","loss":3.353192517783011,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"656648a88aa028250c62d4ab104f452b0834fd2496d04c8ed828a118942e5038","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"7e9984ed37dfd5730fac1c3587044e7797a293241b6964f4c0e4161a06830e9c","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"93c6e2155bba38d5b9d667475178f62901486aadda27e417d13b97a27a3ba540","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"ddb992b2ea44584854771a31ec8b23f0ae96a4ea433510aa998d6a93b0e9a8d1","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"265b072b1089bf09cb26741f795c4e843fd312060079dd77b2309bf1a1485cac","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"e957609dfcba7a7f157fc8c669e41e7aaf07cc6707056c8e4493986c4996dabf","bytes":5204}},"requestSha256":"1b832706d3427a9ad8336984b5a37d23fe98591a0a004c0a099292886ad5489e","metricsSha256":"e3929f4079d14793105a4cdae1d094c59582fd9a1e9ed2de05cfc3bf986cf72c","continuationSha256":"b7abcfaa25c217fbffa453aeb98b3b82e4b3deedbfa68c4f52b6d251b5e4a1d8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"265b072b1089bf09cb26741f795c4e843fd312060079dd77b2309bf1a1485cac","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"eb3e3dddfcf6473568d2ca8b2631298d1d7b216527614eaa8f8105e73fdc24aa","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b72ab8427c791c09952459ad6d404041f966d43c49a0976833474c8b17e6864d","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d83bd101187d23ee2c530a48d81c9e6473d07a4c3c48783a6129b33fb0b4337d","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2461549","recipe_identity":"247aa962ed21e97713a7560db7217e8936a7bff4ca0f52fedd30c1890b6b47bc","loss":3.3573516337837934,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"daaf6960075dcb6d4daa19432598a2dfe07bc4b5030ecdcd8f7be02f97a2fdc7","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"b0901858bb5db0b6b934a25818a995b3187c39b4d79260053ec6e44148affd12","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"48056f3c17d8da92ea52abe357c48d6ec5afee84afc8f2d36c21db2aa29bbe1e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"533b62c0762a3b54f2fc03cbaf90abdb8ad44801520e7812b17fe37a26fb95a8","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"5e7ce16cc9594669ea141fa3f2d9390f792ddeb35e427ac5a0fc540a5cbc56e9","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"b5cc7340542635a9caa69d1e7133b0506cc1de9635f792ebabdf3d8398d666f9","bytes":5118}},"requestSha256":"9d2dea47d91523aada3d0cea86a1420dcff6d3440922eed68cae6636abc19e2a","metricsSha256":"576eee96d588d2a64e4138cbf3ca645290e9ed17bd39af0f95da92faa3cf6eec","continuationSha256":"55ca13e88c69e738440f05f7b84c1c55f7493a0cf32525efd3138de8e2ab7700","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"5e7ce16cc9594669ea141fa3f2d9390f792ddeb35e427ac5a0fc540a5cbc56e9","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"51b1a8c67d9ecfa724cf370a1d0e90231c109ccde2717d1c966de4aaa871b725","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ef9986b011c4f0c1564defb1e51b93652038550b3cd8cc3f15ff5dd6003dc94a","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"247aa962ed21e97713a7560db7217e8936a7bff4ca0f52fedd30c1890b6b47bc","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2461550","recipe_identity":"6de3e6db8edada0c1e424c37fde5f1bd4367288f3194346b7c1ea451a9d81654","loss":3.356640648651833,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"17a6676f57eacff50649a468824f9bbc863339ca3de4e98f4c599ebdb2bfda87","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"473c2b3322edd8a38431f4dd7e6fd7d20870b68418ec08228ca35c999e563839","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"04e5011a4959327c79e448847cf1fb8cfe0536667ec7c9f76c4742dc08bdd70c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"016d87f62dc132e35e5ba072ee1c56d3f1169fb4f2ee8bb7311e985585eb7f89","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"15357c6b7f18eca6928a433540d26210706cd1fb2799fea095d75e9f840853ad","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"1749aecafd1d777f17e47a2e2fc53dbfda21e1e0d72803b6aa2ac44e9a53611f","bytes":5120}},"requestSha256":"e39a13cfe7756e2cf8df4ff5e529e7532f61e28f3c726ab9044ada998f2774f4","metricsSha256":"5f7ee6f96bb57852be7c0b0b8b8d06ea2fe08d0ecdbe37e947566602d0664398","continuationSha256":"fcc5932a0cb00022ccd86e5337e6b71f5ea654e5dcc356544c23007ad4ce6019","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"15357c6b7f18eca6928a433540d26210706cd1fb2799fea095d75e9f840853ad","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"3747fa574918951a8cb2fe365c9e406ac1fe38803d9991033c8d48c1ffe80451","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"550dac5eeb931c2fe5f59da55409dcbead9999b24264ed86885ce5aaf34347a5","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6de3e6db8edada0c1e424c37fde5f1bd4367288f3194346b7c1ea451a9d81654","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2461551","recipe_identity":"21a390b5b7040f7ead6db06e2dda44050a69638169b0a62e0db960793e24a1bb","loss":3.350406513061352,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"47592b1348b1c4f3070a19081c784008d152111e5dbf0fd7286337bee4d18343","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"6c102ba941bbac2be711380081b6e3081c7090e7c18121e82566447e956950e1","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"7f2a04930534cf9d2559c5a14eca9a6af0e3d1d13d95b0195d4c6f60aa6a5336","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"67f2dab5e95eeb8c0b44416ff02eb65520c17e102e93199dab75bbc9ec151d1a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"184cdb61f456a97c72290236f5e6404914d653227e1dc2eca2807e4c804ac55e","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"fe606325ae2be01ff0668e5b7d14b2d2502274ce0c2503cd3560465e63455c06","bytes":5177}},"requestSha256":"4ecae46da589b5d8a0c03e2852ffbe3227d844ea0ad962dbaf618242b7a6f6a6","metricsSha256":"9f2d1496f1468392325c086965bb739d036f588d7b02ae752250918f83326257","continuationSha256":"ea1fabec462891ec5cc6667b83e594b85405e65c16aa4ad21481b0cfe60e48b1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"184cdb61f456a97c72290236f5e6404914d653227e1dc2eca2807e4c804ac55e","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"c731b01881e5b821aa646c3103f8d3f2e30cd148f391c22b49777b6486e81e8c","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"001422b547c59960507e6eec2039a76ca445fbda0f74f6dca72d8fceb17e6970","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"21a390b5b7040f7ead6db06e2dda44050a69638169b0a62e0db960793e24a1bb","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2461552","recipe_identity":"79e7048281e5d51fbff5c98b29e57ce6052e170049fc2d31d6fe609447dbb200","loss":3.3560097341079382,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"480e117ae10486d02f534f9975b79603d29b3a37cbbdcf6ee9a01140b33c0e9f","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"32b775643c5cad304010ef5676c11f349a87c76a085e5088356545d9e7284eb8","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"a32dda4782cb49331d07c3c779b0c7287121f91a980b49fccebbfe29ab730bc2","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"215ec99e934fe9cc342f8e9a695a84a498aa8457c3cdf373d39ec694fdffbe01","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"2f37016fc9c81ae38e5a448d645ab0c8e2d39060451fedcd9001bcd1e04b5ff3","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"23c5218025be8149e721831af37e53c4d838048ca53e4984dce48ff48a662217","bytes":5206}},"requestSha256":"2632521eae10acee6218818c7bd66591f52d0b7b2aaaed98f1f3ee6325417392","metricsSha256":"05be5f99fe8de52e3841b6a9a6d86d0637f2a69d5c7763080343b001f72c7dd9","continuationSha256":"db4103972931918b3a2a7013811e81045008be50704a03c1926c4bba8cb79256","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"2f37016fc9c81ae38e5a448d645ab0c8e2d39060451fedcd9001bcd1e04b5ff3","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"f7b57c500fdbe2c347ab6e5477333b9023857009f8124c84146663b806192504","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"34bf5bbe8b405ad509a268bd545cb8d310c46b8919cc0e74ac24fe98791c57cc","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"79e7048281e5d51fbff5c98b29e57ce6052e170049fc2d31d6fe609447dbb200","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2461553","recipe_identity":"5d9de60c4339178f8dba4408fff8face1812b8349e9ee188c7f567cb1772b615","loss":3.3505589714755843,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"1f5de6f6003d7f83b176e75798526dd26343f014572e7e5c82c4ff7b0a6b1062","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"483f3f1538b02ff9d590f7e008f75ddc679267587190e159600fd61f9ba2c384","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"2776a5d10f958d30416c17064d264ef47bf9bb1b6ef6fa6b10651db49430fecd","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"d502cd29573f5954b4ef69fefe72ebd582d0453a3d40131a6b93d8a8cb6778dd","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"10302acdd3fd783a8ea952a07f3e4939e9444729ed2fc96cefed4cbbf6b1ca47","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"16dc60646fd64f09e4889fde26d383fb3819db12e34d14657f506ae2c6a1feda","bytes":5206}},"requestSha256":"c7f3a63f29df74f60f57ab075eab0e36eb36511f56dfcd125197b83987f15b7c","metricsSha256":"4fc321035da21b30d9684fa1422cb436df56d5b92b41eaf6aba3e412baa98df1","continuationSha256":"d2286640f91ce3ae3e2087a9ce8eb8f2eef5ce2f8564bc0268bf0326a54bc7e4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"10302acdd3fd783a8ea952a07f3e4939e9444729ed2fc96cefed4cbbf6b1ca47","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"4266d1d8cb14fe3840ae62150111633da1cb1f0e94901a19e9c9958d94261ed0","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8bc1f7d21bdc5ec83dd4becb872766371768dd18aba2ad4c577e5b51f633f3ae","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5d9de60c4339178f8dba4408fff8face1812b8349e9ee188c7f567cb1772b615","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2461554","recipe_identity":"1de442cfa5ee313630365e69acc0df9f29d10a1495446caa77d00d874b5dc872","loss":3.3500551367449667,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"f9a9d262c2976c9d8397dae9bb51b32e24f35559c7d54cafa3fdaba50af3d04c","bytes":749},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"f5da36ac909d20ac9ef99aad7a43e04f0301cd3eaf5d00ca4f8d51666d5d7066","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"2db6f315fbe50134ba761d6d647adc55b7d28d78e27a8fd62d1af403456efcf5","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"21b75df7fe9fac49eec82fff29763e055926debe7b72aebd571830a0b81922dc","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"442fe4c316b15517ca582f6914e6ef41f04b7c781d2503162c64b4e9fdb5962c","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"2d2b8feb777c99feefcbed30cc7f719d266842ce7a277fa6d937d27350937b4b","bytes":5177}},"requestSha256":"ede0192eed1acb200c3887d788c0629d4d0d9e04e61560604943a11ccb0d2495","metricsSha256":"03bdc48c8e2153427d59ed0a27ed8b299eb492f78bb06b9f0d75f6f11de5ebb1","continuationSha256":"b7d50c9149591c9410a13996a8796d49b19df0b3f1d48644be071b73573742ae","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"442fe4c316b15517ca582f6914e6ef41f04b7c781d2503162c64b4e9fdb5962c","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"5a4380108244f8655991fb41b6be78ebfb221e1284b25c2ddd1ba3b0ecf740af","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"a479553590175c90429bd0455b4e23ab00c7e9f2a9f52469930c29fcf058fad2","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1de442cfa5ee313630365e69acc0df9f29d10a1495446caa77d00d874b5dc872","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2461555","recipe_identity":"fb69e174dda123e8dce12f0ed96c3f9773f75894f42825bd9327097db7b0f9e0","loss":3.350563310440662,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"23844dca4f027248133f798665a2bc8869f9050a38dead28c8142c2451807805","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"4f1a06c187b8bee4977ca2d92e7be46f4a1a19a6e2bd0c05f891c8e0c5ed9134","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"948da7d4b579139150b25385436e66d467fec6b8c7ebd1b678da5c926f23d37a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"da9ece2d553a5772ab7810035c5f07a10091b9bcbbc6a2a7ec4ec06964f394e6","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"f5147e5b8ea3268f9d3f59d97f1b725b8aff5b6f121a96ce69b020df8a353a1a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"c6230a0e0b2ceb21e244cc6ec1f61652ea73bc1497ba15f6bab1b4a68375a1e6","bytes":5204}},"requestSha256":"a06f41b73fb77c981969bab58fde0009ce34bdcb66932e3b025987227cba5ef4","metricsSha256":"919983370c3464042b8e1c0ce16dbfcf4d5d34e63ca74ad9015a289ccb053f1e","continuationSha256":"b328958f6bdf60f70f542f705d5d9628bfc1eea0b2700f158bd5ae4861de7f23","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"f5147e5b8ea3268f9d3f59d97f1b725b8aff5b6f121a96ce69b020df8a353a1a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"697cba8b1b40848a1bc6a6702780df2f45543c2eee7d6dc2f54f48ce662beeb8","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"27980fc7e980591d2d83efb3e531a8f0c2adf69841bda00affa56d30d16cc488","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fb69e174dda123e8dce12f0ed96c3f9773f75894f42825bd9327097db7b0f9e0","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2461556","recipe_identity":"851ff4781d03ef4df9a3d6e9de4a0af53afd894828008e2909a355f8507f5095","loss":3.353615725642847,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"2c2222527b0738f8fd5c93ff56b8e45623d3916bae4292f111b6fd1622d6114b","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"d96b3ce5fd97542e1f4898c17b6e10fafdfa5c2b61d6dc871265af558e362660","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"7b27c835e03f751f936e01bb3b7df3e37f737956391e9b0a7ff511303bcb4c0d","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"bc9296d7dedd5eee5fab76a8746f535ea00b27d4852e570f50690fdbbd1a09b2","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"fdbcebcc626f0cb8bc560d20b930f2724823fc876c9b234d823dcf7ec41aebde","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"fa1242a2ea9e8a2fcd1f350fd03ff544ad9af11c2399bdd9254d79c68ca14749","bytes":5176}},"requestSha256":"30484abea30bd9acb079d10629d2c58c37ea5c7d925f691580e47eda7754b4be","metricsSha256":"819c78360338c776aa099373eb2e21e1fb92f185c19dd83618bc25e278718ced","continuationSha256":"dd359eec9e92f2d786b04f243ef1774e7537f57a5ba62d93278b74da4b384228","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"fdbcebcc626f0cb8bc560d20b930f2724823fc876c9b234d823dcf7ec41aebde","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"d8eae94d6ea6f14c5c66b1668d7eab25374a290d84e3e3c3f3c542aab8fc9411","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d8a303b6c8b5420ce0f369c110a74ec463a381fbc4d4e455efb904d9c3cede2f","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"851ff4781d03ef4df9a3d6e9de4a0af53afd894828008e2909a355f8507f5095","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2461557","recipe_identity":"9324a71daee969af51a0e006c6c18a308ebc49d9d426e66d2b364dc83500f4b8","loss":3.3500319262251748,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"cc45b6138864ca6283536d1c2aa0b56033c625cd3fd67cfbdbdcff1c88954f34","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"55c56618fac97414a20b3565a0d8c8aadeab6b6de3383acda70403416f3248aa","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"37c0bbe9d3bca6c93341cca62434823f2bb658595eed82a7839c047e4cc8e8ec","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"356734227ce75c6979083e63327a3eb36a0638aaed268034fba92eede0589458","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"32a29005095b698807c09fe3d70be1b85e64c3935caa93456ea9dcc0385aafc0","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"b450c1838db839596859bd0c8833f2e542abaf59b9d4f6ad7d23216483757655","bytes":5120}},"requestSha256":"308abad17aec36adef5274f86ac911d9bafdefc30fdce4ad92a264534c3771e4","metricsSha256":"d6977bae61c186286bf5e6dbbae53e8d1c31abd4c0260444cf14fc52ed9ce6d8","continuationSha256":"29504ecd29b98b2a8b0fd33710766f8849c83c69297caad9b79dad87b1b1573a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"32a29005095b698807c09fe3d70be1b85e64c3935caa93456ea9dcc0385aafc0","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"495ce7488e4d7a07d245587c65fa169e9731670cf6bec9a464fa84548e0d6919","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3b657e3115ac0f8e09be03cef695c271750ea0e401dc2b6f69ee254fba77d5c5","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9324a71daee969af51a0e006c6c18a308ebc49d9d426e66d2b364dc83500f4b8","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2461558","recipe_identity":"b927c62af4c3b1d479bc828d23b03c370c13a90ae9d07aefd5743ff07e4a1122","loss":3.3538007354347514,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"2e92c4dcfd4895f6f05e21b3b259cf7bc802a3f0412f489540cdf4d48f824802","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"c56d601c00890f047e080b23d192256e072bdc09540bdcf05f2e232eb225b27d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"548690ee9c66851080ad5606e36a54a101b6ca2d1056aa6437f2321d1d973cf7","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"8ff7fa2d0c5abb9d794d29344f62dccfe0e5f93e70e5e5aca50261acaff5eb2a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"93023cefc6b34ca11322b5f364585b19a07b121162a4ff89f72fe692ac22b2d4","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"6f87c55d24420486c16ced9b68b39ad5978a80ca2fe0a1fe09f4feffc5c27551","bytes":5206}},"requestSha256":"f1ae055fc4641951b8be744d61d71dc80aaca515be58c159933fcd40d0549edd","metricsSha256":"297218da006d7b54758a7cf0d3ef0933513eb974070dc596d6dcf5aaf5280c8f","continuationSha256":"d1fb6f7120d5f7ba45329449b02e46ad00560f6e38fc0eb1442a76e3c62edee1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"93023cefc6b34ca11322b5f364585b19a07b121162a4ff89f72fe692ac22b2d4","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"1d587f0fc78422d6193fb912857f7db9c349b43b2a4b592b788093fcd42a2c73","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8d22563491db6a416885f837b83d89b37c97bfaf3250019121820fdc197fd480","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b927c62af4c3b1d479bc828d23b03c370c13a90ae9d07aefd5743ff07e4a1122","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2461559","recipe_identity":"f5bdffede29907151712f3d116494eb8738fbf9aa71abd412c97e6c45712b66c","loss":3.347097920415326,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"92d349896562bd2eba03996b157d16a5721d51bdd03fbe71e4bb78064dc8c5ac","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"facde0a8a572f91744df3a6fc6bf5eaee254077866f0d1b4c4ea3161b3f89c5b","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"f768908f87be683bc7ed3e2cc3cb85c2a48c6fecb2b61f61223310b97ebee331","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"1181987151d60171196bab1884ee2da68ee2626fa973574e011e7e7e7afe03bd","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"48ff4204b0f1cf2829f2b61c23fe28b8239f4575561c0fa82b33a9b0a28a846e","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"3aaae8558d33122a51310917e2c17a013dc13060b4c176b38eda4fe446ff9ed2","bytes":5120}},"requestSha256":"38f87d358502b34bbc114973465dc10851cc51248d0b06da3f96974b8418e790","metricsSha256":"68747eb4f1c4689583207bde6245975f671e45e0994a537ad591a86f7a0022d4","continuationSha256":"3a831ba6e0424059261c495378cc4a2230e9046da62e2e03ced820a050336b20","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"48ff4204b0f1cf2829f2b61c23fe28b8239f4575561c0fa82b33a9b0a28a846e","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"503c27c64ef76bcbde2cc824d74ce8e933c07242c4f450ccf73385935b914530","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5dfc9c58461fa095ae4664bbae626305214a7ce854b4e4378e8bd94c48e1a24f","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f5bdffede29907151712f3d116494eb8738fbf9aa71abd412c97e6c45712b66c","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2461560","recipe_identity":"86f77c96222b09e31869ab0b4ecf0dd889917f31bb3a90bbf8a735fbb4a9cd11","loss":3.3516200603362734,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"e087d99bd591fd5bd30656264592a95d0cf6a761373d587bc9d097b479fedafb","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"c36ac5a26f3dbcc8558e6617cd3b665710f8b88519cbc5f786a574a53b8ac8c7","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"3511c1d58baa7cdd4a2af22d492669e98514436caceec40df7cbf6b388495692","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"0fb17df5588e46c139e569bfa4816d8055cead8cd1589fa9da191914fc68e91a","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"33dffed2c89a5b68a478a47bb997a46d81771b4a7d83dbf668d720275c04fef0","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"5dfaf8b97cf9fb5758d55df5d994caa764af097bb3c9a30ac93697e472ecdd63","bytes":5118}},"requestSha256":"b25d90571fd407ed14dd93212a2ef774024b534aba33bcfa1c7372e8236cd85f","metricsSha256":"7de7d69b983e3cb59e763477bdae324aed5126b92fd0ba4d94a456ab01f772b0","continuationSha256":"1d632491d13431be44205e621ae837909d3ef75ce9404db1e008bba0ccff6418","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"33dffed2c89a5b68a478a47bb997a46d81771b4a7d83dbf668d720275c04fef0","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"8bb2b434cc8e4c6ec2ccd992be89c53e8f6df461e7c277dbc38df321ca2549c2","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"29ca2b93585cea1c31a0d677cc69cd895985f21be578b47cfa5b4fa05440f109","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"86f77c96222b09e31869ab0b4ecf0dd889917f31bb3a90bbf8a735fbb4a9cd11","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2461561","recipe_identity":"24f65db8a35e4dc278e9bf70200b5f5e9cde065ce3275728f994547d64514669","loss":3.353726081056074,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"a3751a1440edbb5e0316cc2555c48d8c22f1d0dd7cc6ad54ec1f3b59cb0458c9","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"d0031a5e1206573575f07634e18c5105e8c1b9f6a27757392b32701315e86b53","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"e2d3422e87b1db23834f9f6ed91968802b91f8f0126a15afb1beefbbbec629df","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"7e3ee50662ad65c270fda447e1b65e60f2682d1f8b55ab917843ffec74c44709","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"c9123ae875fc791e633e2ea86ee97d97f4fceee583cc381544dcdddec8bbd27f","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"db193ead89f447ae9ec6604d5ffa5db2194bb5c9008e07a085d521ba9fc4ee4c","bytes":5177}},"requestSha256":"223c1ec003ea165d789ee03e75a1ee3625a8e57f7adcbae383c7438029f018a9","metricsSha256":"c6cf98ce29922f5819ecad7ac2c1e0181c88d15d3080697f48396440bb726bcb","continuationSha256":"8fd32d54b199fa22cd6a083da2be1191b467e973d4d7ba04fedf94c512c3cc24","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"c9123ae875fc791e633e2ea86ee97d97f4fceee583cc381544dcdddec8bbd27f","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"8461b76087784823cce4775c23676eccedcbf2b1b05f68f4b9f42f58f3b84317","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1e32e7011609320f54750282acc761e329c9f33c3d20baf200a297c8202c8407","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24f65db8a35e4dc278e9bf70200b5f5e9cde065ce3275728f994547d64514669","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2461562","recipe_identity":"5fd83c340a2af6b98372cbfc076970e0e217376c0f8f6b3d658339387ac44fa1","loss":3.348942577352643,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"2ee7dcf1b389cb1bb602debc4c3e52fbaa4f70626107f6498c6238ff9e79ca94","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"1946b5d8d24e13c6f19afe40e341a38bd639d13d2311370c6447f976aec8bb19","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"dbc28a52f3b5a922e5c424e3c1a05171ef474dbae028c6ed4597917738940187","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"5dec0fdb06c400d7aba64c8c8cfa3ad8bdcf4d196261608a108a3e99a6115c27","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c611e5391bbe49effe5a03de05bc5165694e91a4f033ec39bd0ff2171fd0520c","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"b577b6778e0e15531a8d51e50405f8ffe11ff1b0bd7562a046b7f073e4318e41","bytes":5206}},"requestSha256":"2b1129d69a95e061380d00c368483fa6cff6ed82369a34018a1f7a307e597016","metricsSha256":"fa40caa1b15ff9caea4d3bda832041848b609e88a98cf158c43fbebae8165f07","continuationSha256":"bb7b6bb970e96d1c5f5afd989872b33c1fa1af2ec2e1414d969c3e0d41e691db","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c611e5391bbe49effe5a03de05bc5165694e91a4f033ec39bd0ff2171fd0520c","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"0a94f1aec2ccc7c59a63768607d5fe9717490450c52a2aff4daa54726aa47911","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e3bf5dc98f3c46c45ac05f37d34fadd30149aac5e05ffaac7af7875905aca867","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5fd83c340a2af6b98372cbfc076970e0e217376c0f8f6b3d658339387ac44fa1","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2461563","recipe_identity":"408ae6b36e36352c63407fa97274e1a8f2fc1146de32a9cbef9c4e1292d30021","loss":3.352916446734371,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"47e87c981987d4d9c749c1849e2c3fe9214606783a64c748249fd67a90fba9ef","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"4d7e4a1f8415c3e406861394db9b078c4aa54f5b3cf69dbabc0f667a212c3101","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"df787a98651a09bedfe6109012eb0a3acaf567ccf5bdfe1c331ac9679d7c9dd3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"98a1211fe6e602db808c1b177b4a29f945af44b9022da9ba3b7da114c5605f27","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"e9931246718e06187eb366bd60e5b5257a1671d9b9b98cd9bf939bd24cf27ea7","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"3d58fbfc83250a63ae18465207da1f5d80f7aa21b53adf87d7a8602b39ed7783","bytes":5120}},"requestSha256":"431ce0dcab864e2b8e14e508a7d3d1a5a905d9654ed86dbdfdef1e813d8ea3ac","metricsSha256":"91510cebf467656f74d6092fddddf32a3be9b02394cfe78e612f1dfba6eef8f3","continuationSha256":"e2896706029eb410a245304875f5c65de5772b4b117f4bc8950d93b76f055b76","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"e9931246718e06187eb366bd60e5b5257a1671d9b9b98cd9bf939bd24cf27ea7","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"d7810c139765d51527e0ea8094a0227980950a0f2e954179034669ebc5a98980","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6d7daa3f0add95ada140751476fbb446152dbe780ce74d3a968cd6ba89413c6e","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"408ae6b36e36352c63407fa97274e1a8f2fc1146de32a9cbef9c4e1292d30021","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2461564","recipe_identity":"11d66426e6961c59da6f7420074a2c0fca0dd52221ad92773acf0a2f499c4445","loss":3.3537109313600366,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"9ed1afbcd7f5bee2b6218a82cd13b9f2353fc7863f830642c3122d9f485af71f","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"88088f820902218a4cd3f225a575fdf94ce055dd294b05f58cdb18a1aa82cb0d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"3e92637848421e63454696c44c358090969df9b283a7055c4f7848f72d53477c","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"29683d24441865bf2cd3d5c4b4f6db53450bf7100b8a09487bc621ce35dc0ef7","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"4f99d2f4a693684310cb7ea9117d85ddc7e21ed6830f4a50a7d12b32870a70b2","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"9c90219064cd83c621e07d41b0e5407ba6f9924c84c5b2ed6f88fcd75efa7602","bytes":5206}},"requestSha256":"ec5320bc5cec16a798c60c589baeed9791e154054b259b1f95717b1f7f05f023","metricsSha256":"a19cf1407205c811f1b68b7cbcea63e6fe96a9a76fcbca3e9f76dfc5f89fa906","continuationSha256":"6ca9262eae4a8ff75d3e94087340add7ecfe5cd56926b5eb57da42901c0fd2fd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"4f99d2f4a693684310cb7ea9117d85ddc7e21ed6830f4a50a7d12b32870a70b2","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"ac4a272d1ed3b8c25223473a416cb16f675416fa582f0c365d35e8af70c84ef8","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"57f548fad9b2b8a1fbcf0ee712aecbcdfbf49f5c8a9e8258ba84be7807dda428","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"11d66426e6961c59da6f7420074a2c0fca0dd52221ad92773acf0a2f499c4445","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2461565","recipe_identity":"0df5f00389d8db3faa01e1d498ad5afb378b3a037ca53246167f332986851a1d","loss":3.340056066217504,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"4d239002a1ed87224f16a1f5b3a4f7828bee0b673eb42791a8931043eee9db83","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"c8ad99a1d91e10d9192ded63f561161842cbff77a8a298530ca3098e949fee61","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"76f0f01936f747cbf59c35d1c21249e58557091c99300f187f4f6a3c76e7035a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"0b335161386f977338016f9757990a735b421f344b0da4c16440d89da6e81200","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"8900c0601131d0cb962874d25ead3bed522d8788f32703016e6a6f6efcacdf51","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"f0ed7ff3fa93e8625b7d1312145168b52cc238bc5c40564fee5d2dc6181a65c0","bytes":5119}},"requestSha256":"b5001668e4fc3502b3ff07d47b3f1d7f9224ba881604fc358a3d0cf665893a77","metricsSha256":"cbce6ee76577d2e89ea33bc6e27e217d8552b95cb3a611dc807a30bf76497338","continuationSha256":"4c9f1a53105e40835fada3a4c403120113d5a47ee1edbc9642a11bb2f14f59b2","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"8900c0601131d0cb962874d25ead3bed522d8788f32703016e6a6f6efcacdf51","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"d63b8b1fd3e4c99513a4bef2f41d32888e6f7258628b79006b7fc407b2d5c02f","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"818447c2f231b63d5c0473a2d8bf1492d0a0dcebec675000de052a91d3ddf033","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0df5f00389d8db3faa01e1d498ad5afb378b3a037ca53246167f332986851a1d","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2461566","recipe_identity":"179e6cf55329a778c1fd076e5900699a117963dbd26fbd91da71971f38641593","loss":3.34416150046102,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"d44619271431ad945187b2c05390be09f353c494e5fd0e0ce2a8b72e3ad2fcfd","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"917630992948e97849d5d7a20c386ba1408187741f987cbad4c65a68dda73535","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"a44d9fef1e850c3a0fa3354f1276cbe757662ed71e19b848c2e831c8c119b938","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d46f93b7e8f67646e99fa29b6977b4861bfd34eeb604d1b19afe04fd378c5292","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"4eacfecd7d550f665ae941b957ea36b93c76edee593b413458fe951111fbd134","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"d1e28655e10f5a944c8eaff28d63e21d13ea4e5287b73e08ae50aae7a03c0c1a","bytes":5118}},"requestSha256":"e3d17cf67087f0bd6ab7e9c4fa35848b2bc9423ec7d8bfc5c546d57b6685e7e4","metricsSha256":"67865ce42154111de2f880831692fb3d059726480660cfb5e8398dd4bde77bb8","continuationSha256":"aa5b0ff5a1ae25e1f1ce6050995ee682cbbdda0b7d161fbc51b4f61c9b9c154a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"4eacfecd7d550f665ae941b957ea36b93c76edee593b413458fe951111fbd134","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"e6a551643324b09a8c72d109e22bbfae59d232272d2e0c9895d3859b5a4935fc","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"996d6c7bdf53f9f65e8863e49fd7fb95772e9f3c14d02e5510089380c2122c50","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"179e6cf55329a778c1fd076e5900699a117963dbd26fbd91da71971f38641593","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2461567","recipe_identity":"bfa888f0f8be12dcebdc78f9997c195f29eeb1f6b8ead2fa0c306fe655aedcf8","loss":3.344916610736368,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"a82711d326247276c373d1487282583630f7478cfb761d400b3ee67d2e6a9f5f","bytes":742},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"45685bc81c26236bb09aed96f3d1baee8da0672f3c820a7dce2ace89c17879ae","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"955d73fac9f58027df319d0225b375918053f81baef97d3f2328b668847bbac3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"38febe5c4eb276505b860807cee71c9c556306c07310827784bcdfc7397807f6","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"6bf49cbf150b7f0030209f45d654d238b78fe484ae5d09e441d76f5c4c78567e","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"3d80f21f3fa3ad97d3b09a705ba7235f91b2bd6394ad7a9fb1277bc93a5aca4d","bytes":5118}},"requestSha256":"9785164003d52e1a3bb73448a7f24803f06e17fd8fe679d23a9bf687da49251c","metricsSha256":"20051896dddec2bf96582cacec5e397a32dbe7633210a671591e46d9433c4162","continuationSha256":"375a280d8da137339e2708c367a414164264b782da3d55e17e59501a3fa7ceb2","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"6bf49cbf150b7f0030209f45d654d238b78fe484ae5d09e441d76f5c4c78567e","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"55c7ac2074e536ce81e601a2fa89a18e2ac88d26d73c137d71eb2a9e89890778","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"17ec99c05e99be6edd20011c1b70454181f2c313f6fa39f9611c24a659373764","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"bfa888f0f8be12dcebdc78f9997c195f29eeb1f6b8ead2fa0c306fe655aedcf8","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2461568","recipe_identity":"ca74eeffa1309f84eae15331922196dfe7300b4dce96c52ae4e7fed76ad97a0e","loss":3.3428652603456044,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"b5d8ab650cfd9ec0301934c7c6b62bf8d58b2cb018da8975a11415dae3785531","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"afd506ae67de1f594d5863e40103a7faf7aa02978f8b46a9459a9c1fd83cbef8","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"1e03232751fc2dea837cc92e817c9fc8e351dbfd820875878ed51e516e778fb9","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"7ca2bd3fd0909ec3dc7ba63a74927a7df1bd1b732670665345dd0a320e192226","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"73ffbb2facc4ece9d2280ec628f5cf9bf6ac6e93ef4698e7fdaf4e9ebcdac56c","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"8310fe59f62225bc49e15676f3ee8678a21c3f998e405cdf85729aa321a56718","bytes":5177}},"requestSha256":"2fd0e178038cbea4e7ad5698229a441779462534bdf0779e452f0882cacb4924","metricsSha256":"81336aa56983771535e41f794f56a79da381f6da80bbb7a1295b7cd2caca309a","continuationSha256":"f8776135a4bc781d63936ca507d116da13f9e56588e813a499ee5076ee45f48e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"73ffbb2facc4ece9d2280ec628f5cf9bf6ac6e93ef4698e7fdaf4e9ebcdac56c","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"6d876cee114c562d8dd446afc75cabe06e02f8ea5379b95feca3f3bf9bd45988","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"46fbf7a7cf74bad757dcf7a92e540de95a04c73d3bc141f8141c1e15488d6961","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ca74eeffa1309f84eae15331922196dfe7300b4dce96c52ae4e7fed76ad97a0e","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2461569","recipe_identity":"e34d1a33d6f759d6f390c5b6ef53ee349cdd9f95abbc42a03810a61a7d18ecc6","loss":3.351132205130257,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"7396a71491416c237d9f6d4b42a4502ede93e822dcbfccfdec2758790b7d7e9a","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"a073e11ac12ddaa35776772538324efd9a56d4d359abab2b13a99b5e62131a4a","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"98ec9df97eeca6723f8953d7a8a962c11061b9241d71f3321a84a1a15020aca9","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"e4c8ee5bae2d0e224fd9a89e7708c71f30f66e7a547a8e978d3faaeab25c0548","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"c369153198d4754c229ca334a5fdb25d1183e38ad6d20df21a4392134dad0654","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"6e1f8e26993e0e152783752074ed24166e3b1b96efdfc608c0214a76a1ce17cd","bytes":5204}},"requestSha256":"809877858b4b061d1ae18320e3ed5c7c5f101b4134a0dbfeb1306a383f6bd7bf","metricsSha256":"4566dc98abc028477e209a2f4b58075ff3a1ad80419895153c16a4e27e3a5053","continuationSha256":"ff565752973415122cfb7de7979a1b4cbfacaae0bff0d95ea0f584918b3c8b3a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"c369153198d4754c229ca334a5fdb25d1183e38ad6d20df21a4392134dad0654","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"c7d2e33cc6dada19d9d892976fa3fd340294cabe0bbef1adb4d0aa7800f6bc0f","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"11f5d721c9373ef666d9ef665b9f7e553eac1bb554271f3f4550e0642e1bdc5f","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e34d1a33d6f759d6f390c5b6ef53ee349cdd9f95abbc42a03810a61a7d18ecc6","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2461570","recipe_identity":"d11cd4a00bda5016533ea67044ec83a768d288ef2b0763c834bc0ed5ae0c399b","loss":3.3427909079978417,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"74edca8ce2d8969e0b1c47312d24c86666ca7a31540dee40ab430d6e8f17cdb6","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"db7ce1eacd3ebbbd1f07aff347b74d2c1bcd05c8c65ce00847441245e78d275e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"9cfa565b01d0f4fbdbd05327038fced8796ea100223e505f1649f049c0b6c895","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"c34f17d0cc73b029caaec65656a2f2b68a0eef87587d529d260d64e2b5aff220","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"cd851a4cff6d42ced1545681535b5c5ffe583adf49124d929a471efe5f6d09b4","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"b33b22d87b3cdabe925b45286bfdff3d81a6dda3022c152c73a35af0f855bb1d","bytes":5113}},"requestSha256":"098d31f462616ac53c09d2d4143c4ddb2b164a52af332202922c3c061590bee0","metricsSha256":"993f17b81113d7bb3af1ed787591960f8acb4daa4c7a1f94c3fd20f66345b24d","continuationSha256":"5462fa6d0d53edeffea707afd88200a6f6b70ec2008cf6c7a6e464fad268eb0c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"cd851a4cff6d42ced1545681535b5c5ffe583adf49124d929a471efe5f6d09b4","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"b01a22948605fb8903febf8ccaa71950fbd561e9c9137e2659bd7dcaab665aab","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0ec87355cbe8f4d0d2c44abd0a376622ab52dd7ba8fb547e521ff86f97b02e99","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d11cd4a00bda5016533ea67044ec83a768d288ef2b0763c834bc0ed5ae0c399b","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2461571","recipe_identity":"b7c465db5c2e8407b878538541c8b36ac874bd9de262b682d4a2bc1376a6834a","loss":3.3358776806003805,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"f4bd6ebef20468ddccc997e5bb75c58b5c3292d43c237e27f223dc65fd3aaea5","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"a8556111cf6e4b5e3891b1339261ed21c36bd4b32b526f0fa73af690faa8897c","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"1c7ab54bcc89abadd7d42915e73de76de31fa5c8611c6a49ab52a0e90dd01f95","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"06ffc82837014e49a40c70a45872ca28abb3b659bb3cc140fa760b7628ec3b34","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"ee0ce6846348d22804d239fedd2f04aa55b72785af48e8fe6efac42e1cba2318","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"fdac57abe64a584a2e326704710034720baba0a49d9a1a3ecfa24e587cf78b35","bytes":5199}},"requestSha256":"480c593af1c069f62e8f06278861e855abfc6af1eda95955094f427f0f0b2dc6","metricsSha256":"4805c77c9e4f0fb0c81db07ecab6ec1c97e94245bd28f6fd16e0edb2bed01272","continuationSha256":"3ee9ba81715c1c24f86df954c2f40f13008cd50257ef51712fafcced7015b478","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"ee0ce6846348d22804d239fedd2f04aa55b72785af48e8fe6efac42e1cba2318","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"b9b6ff1a484f053006e19cd38b0321dec12e7fd72be54a4ac2ae1cf1f5b0a148","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9fa14213ce387d491e6b37eab5bde48ded0bec1e48b4aa75b3b259a2e6e877f8","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b7c465db5c2e8407b878538541c8b36ac874bd9de262b682d4a2bc1376a6834a","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2461572","recipe_identity":"5ae69c1395311365e60244b94ec1edad2f1d9e1795234f0cf2478b8af5dcc180","loss":3.342626461286864,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"0b9c7cc6b0f09f870546f4c48c28c81323bfd2857000041b8cf52c0f50e48649","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"fda64dc7237eed09af030923eff7a5b0bfb68ea78777a4259456b3b7280e442b","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"997f3f8baf5a4b881bc65407ea76bacfea4af781196321b6590bf3ec03e8330e","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d711c1838ca9a63423f63070483f1cf7dca443e386ad7a6848e9bdefb856bd66","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"e171486d4bfa4d6ca1e7bf693348d6d7a882603877a51a5c0fd21fdc7c18f61e","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"644aaa1503bb8d849191c2e4b065120e455ae6e723dd2eb9beaea0ab8c13a2c2","bytes":5176}},"requestSha256":"e21b82bd4e8cdeedd45959451a232b42af88a6ec95aaf1110e60068ca0492542","metricsSha256":"4030efe6f1b612dc4df463f084fa6639f9cdbb698bd9e846da72ca6a825cc55e","continuationSha256":"712ade05df490e24aa6390c8d1677c59c0cee7582f13b103972e8dc8d1fd5662","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"e171486d4bfa4d6ca1e7bf693348d6d7a882603877a51a5c0fd21fdc7c18f61e","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"eabcbb13e3c95f19c3140995e31abf0b35dee8347b244533fad13d0a591cb7e6","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"082cdd3da7c4895e3529d6514715e4efa016c70241bd47bab2bf90298d8a98fd","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5ae69c1395311365e60244b94ec1edad2f1d9e1795234f0cf2478b8af5dcc180","loader.blocks":2391040}}},"sync-wsd-h120--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2461573","recipe_identity":"84c3092ed4e4a0c32ec5f2f1b7c260bfc60da483835f5dca951fc5d86fed2299","loss":3.3389874721956683,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"b99b078d902e89ade843afca76bd923624ea4f54847bf5911906cf899e1022f8","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"c506c34553bee9a720b538bf50507a76f87257f1748b0502b92531c4446e3894","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"2b4d4aa233324d03f47fb05ce8a1e7f2b90d00981b416b5cf39675c1ecbc3855","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"1968e2b0961565a8dfd4bfc265e36b9a3bcea8636496b97a8f7d0030415dc731","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"89c936a2ed865e943ae8a64c0c72f2c631fced0cb3d5b1ea135229e604d50ec3","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"c84fd40debdcd5ba23c9db0460de179c2f5587167e75744b973aec45e2541962","bytes":5117}},"requestSha256":"3f57fd8b57187305463ad4277185a40f19593e9ed4c27555bb5395953cd0e50b","metricsSha256":"c12d6013b74169f6c8699d7025a72608b28c7b035ac0bdd3d182152c266d92a2","continuationSha256":"60e4b7c949118c22b10d6ab35c3f8a2f5b61bb6b4db03398267be8d1705a2db1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"89c936a2ed865e943ae8a64c0c72f2c631fced0cb3d5b1ea135229e604d50ec3","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"a3a2c82d5e70de0fa3865202085973debc3d1013cbadfac0e64358fc34ce689c","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"09b98b23bff6095db3084f2b7820643bf7eb063817e7948414c7cb57947891cb","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"84c3092ed4e4a0c32ec5f2f1b7c260bfc60da483835f5dca951fc5d86fed2299","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2461574","recipe_identity":"d2f482a3e2d0182a43255ef0a86e2682d60bfe98a21ec1b4c32d4d7ca8c2bdb8","loss":3.3372535258342455,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"6256675a4f4c8076796374c9d8c894bbd69def6c41f0062fd9701d31dad261d7","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"883b23796851e69433956d5e6db6c40f316f226d44a75aaa51b37c4f610108e4","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"c18bd3e6a547a1f19ffcf25ec881f7d7fd5215b5ec27f9f3cba7cd302e3d75e3","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"3c4103b98707b20acc1ad22d3ff51afcef6c8bda293f74ef1369c5550aad60eb","bytes":244950123},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"49878e8aa91cd34d5556eae2dba7daf54282b299b70fcb2af90a77ca1d7108ad","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"e359ea5e51af020c5e34af24b304f4f1bf04415f9630f5d45b38a3a5aed9e7d9","bytes":5203}},"requestSha256":"ae5a93105f055943bd71512537df6396c594f07187508efc9f0644298af385ca","metricsSha256":"2640c3a656f5d0b26a9631ca99e3f80ef0fa4578d5925c4f7bdb995fb35efc34","continuationSha256":"4c5347695975cf430c8a1b4acc9639be5936e77030ff9d62fa57a8648be23739","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"49878e8aa91cd34d5556eae2dba7daf54282b299b70fcb2af90a77ca1d7108ad","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"283c971f241a2a6a0692bbfec970b890c1802fe73c551f419a970a174d94914a","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"4acf9a5ac805fe1bc856d5c46fde88fd028d15646353cd7060f587432d3cf68b","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d2f482a3e2d0182a43255ef0a86e2682d60bfe98a21ec1b4c32d4d7ca8c2bdb8","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2461575","recipe_identity":"42a550ef7dc3bf3186f508f9727fb9647c7924ad44e936583beae470cf9e964e","loss":3.344210140111933,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"9bfac4a1dd1994d6bff50c57ca1dba69a4c5b0c76ad95fab2529bfee53cfa7eb","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"66f930943baaff6292972be88123135f8e6a680232b0db19fb2cdbc810dff654","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"004693c573696df6cfe699ff3c3b052a971a15177906c1eaf4472ebed0d3123a","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"7d5b87ab5f6e4b63e9593dcf9c97f760bb3518b16a427adffb3d10c5d1dae44f","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"3e1671d203a5ae8a0d57af1efdf8a36445cf3515a2a606f6c7b821b0cf649b61","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"08e89ba012bf9f40e801053815385cdb31dcd20853f471f6ea61a3e5cc25c896","bytes":5120}},"requestSha256":"ec3a1a06a686b3324b3d68f2d5e7682876ef19fe19a1aea5a58ba332d2c9ca81","metricsSha256":"673e6740134017b728fd39b7880d9686f68884c07eb400a122b1e6330286ca29","continuationSha256":"d5688f9b777faf390c142bd49dbcb67c8b8f0623c5df179dc4dec51bbe88bf6f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"3e1671d203a5ae8a0d57af1efdf8a36445cf3515a2a606f6c7b821b0cf649b61","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"fc062b64ed944f7b3aeb7a597ca7377f41cd66413d25f9c57bfb934ef2e70904","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"3ca6645273253308dc0974eab53881c87ef9c5d43d5966e7b1636fa0cd0cad3d","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"42a550ef7dc3bf3186f508f9727fb9647c7924ad44e936583beae470cf9e964e","loader.blocks":2391040}}},"sync-wsd-h120--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2461576","recipe_identity":"03a0deac21eb5680d309131f90d4177fa7c69de3bab1a59b0d02e8c2976e024e","loss":3.3385842410922026,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"55078f6275b6075739249eb5c272b5549e62bf30b0ca46b45b6d2d9d22d2c995","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"d8bcb5f56aa9baf6e10f7ad342992e7df725d06166b7e823e4b57d1c8092ffb7","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"a13dc89898a32d4e05934070525c78723f37eb9b3dd08190edc4faa96603b1ee","bytes":81621488},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"6f676a812712c549cd42dcd5f8ea2d2d71eeb5526127823faeb0c2e8f15f71bb","bytes":244950059},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"9cbfc88c46eeae5774c7adaef6e614cba3b57a9359e0fb961f1f8a14b0913a82","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"8ebdd99f1b487315f819db3f92ad692c13f56c707c2245823511066b16eccb45","bytes":5118}},"requestSha256":"fa7c5376a7d001a8379c4a12e0475b962126b4e57cfb4dcfdc501e13abe420c3","metricsSha256":"26ee29d9cc6adabc96c9c703df45a87b47142f1d9d60fb966f06e3c1c011d7ad","continuationSha256":"b4d9ee250ada8041622c8bb2c33f7b80accce2e9bc9de17585c6b10de3e6e853","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"9cbfc88c46eeae5774c7adaef6e614cba3b57a9359e0fb961f1f8a14b0913a82","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"e4b1f6085f50a830d4b8a7770bcd2d078818e724f6a2218ac534f5eb3725832c","bytes":244946027},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"53a4d08086815092f0617bfc5e77dc956b4628bda4c50356d2405f5de4019e67","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"03a0deac21eb5680d309131f90d4177fa7c69de3bab1a59b0d02e8c2976e024e","loader.blocks":2391040}}},"sync-wsd-h160--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2462419","recipe_identity":"a984d2033c5db528d67fa7b73e715c0812cd76b5c967747cf7195bd046d42ee3","loss":3.3454572941441034,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"b6c930c63a3743c22e4aa412275ffb21c77b35a830ccf9a1aca2b899d637fcd2","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"25643e47120709cf0e92f73f61c019e658a172ec765ef4f7e308617c899dd979","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"9844de9d5b07a6f7b8cf5abacb92f1bc6af4e6e3e73e16581aab0a7d4b14ec6a","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"8fb11f02d9c8929f144328b902ff8d057f4677e4744c275bd7c18f417bd55829","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"6df971c9f32894baa692693ec2b4455eaed2ad631266a5bf244ba5e21ee41ab9","bytes":5120}},"requestSha256":"5761d2020183a19a65acf5716d2542a1de847afd114b100c23f48652c887b5d5","metricsSha256":"1e65b9b6e264d9c1ce2ebf1ca1a467d7c458814ed5c9129c202e46e61dfb6b2e","continuationSha256":"ac84ad81c7089635d0cdf7a11fa04775c924dd0c3eed9ffa62069c361a0a6eb4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"8fb11f02d9c8929f144328b902ff8d057f4677e4744c275bd7c18f417bd55829","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"3a26cb33a32bbb13c323de4ec4603ce6adfb42b32be0c72d822e8816d18c5b95","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e98adfe88807b82f846d5985354d418a2109d4dd982a203100d8b0dc2eb2230a","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"a984d2033c5db528d67fa7b73e715c0812cd76b5c967747cf7195bd046d42ee3","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2462420","recipe_identity":"46631432aa9befeb8ed70754a6b2ec4de797c5e98aebc271dc140c771d963543","loss":3.341659362961695,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"6d5dcf89e7d8afc2b5bd3a2edd0a81f7b286363879b610fd42516a618c43de67","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"4d40f5e422324165d2276359d2cf217fbd52c06a7354d337cefd8d1f4f5606eb","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"d74413bfa8d4e1f10a496821c266b13d35524a8a6c2e2945863ce8661e1820d7","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d90dc2b9fc60aed35d1518b5a07fbfe6cb4bb9bbf089ba0c2c6eab62bd9440e6","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"7e0c019aca2ab1e2564809c07cf6756811e85dd4f073d6c69b5031fd29e8e9d4","bytes":5206}},"requestSha256":"5c9b04c9e9cd206d51d76cca6e83534f773269859fecf180bb9e4ab465d7fa86","metricsSha256":"166fd48d2eff74e7c912f6e4f7db169d10388643171488816fb29561610d113e","continuationSha256":"a9cf4fd6b5bc16b41b6b8f5521d0974e9b280ee3d0a1e1e1f1edea23eccc002a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d90dc2b9fc60aed35d1518b5a07fbfe6cb4bb9bbf089ba0c2c6eab62bd9440e6","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"7820553d4fdd11885980f1b9166d20075727fcd5350ee4872c1d580ff5cd131a","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"723e3947164c972bd9a84c0544c97971b9457914078d31a9341cf91dba93203f","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"46631432aa9befeb8ed70754a6b2ec4de797c5e98aebc271dc140c771d963543","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2462421","recipe_identity":"65a49504fa3a36ad6b04ee10e6143ee4feca1c8372f694a6f6b579cce50a0bf5","loss":3.344833484559326,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"ebc3c254aefd01c99eab49fea43260d8b0f9f0af0f9dd467163896fa09b9d3c1","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"01992b5eae16c20b7a4575e490aa8a6daf81bbb13a9fda257c79569261bb9e71","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"668a53350bd8544fe89571e70e50243a44eba7862273d27324fa589c0df6810c","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"3ded072f20d0203378db57fcece5b08e3642f5b55c3cc4c145d2255f3c402be1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"24f08441b4e622ec916cb43ccb96d87352366518a4c0e32e088113d3ce7e797e","bytes":5118}},"requestSha256":"59a470b363d0b29f48417d2cbb5580c1def0e2b05695d033aff1ecdd3e00bf1a","metricsSha256":"2fe347ddb18da1f7ccbbff6dea38a27c531e8878c9d8b613cf7c0bd84a41734e","continuationSha256":"b82470a8dca3e92a3d21235ce97f862b6a5ffd9dacdcd477a296233392a1cbd5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"3ded072f20d0203378db57fcece5b08e3642f5b55c3cc4c145d2255f3c402be1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"403a28e8de2999a2229dee05d1ab723023eefcd850f495b778ad7b721e78e5fe","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f4ba144d1b59a693f5c7de8e60102bcc63839b7d88127850a54f7e370dcab36f","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"65a49504fa3a36ad6b04ee10e6143ee4feca1c8372f694a6f6b579cce50a0bf5","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2462422","recipe_identity":"62572c7f0f9288f3bfc57b95e8228074048c84ceab67e0a6715f36db2c04fccd","loss":3.3511317501123843,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"5739738d76eb06f625ac300dc68c4b955e95315def3a2ea551bcb4a74301a7a0","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"83ac79cd01d60d3151bcc55f33bde8b2a5c7b032aee4df40b2fc7c716bd06c7f","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"6d22957e1f416bff86450f5c67f9f7a2a5a2363e3b751542ab9295292fdb9574","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"9ef80dba6ffe0002f5607cf8c7145dc08746b62f34ff4b7c68fa3ba8988ed41e","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"f0ab7980701290d3d370d608b4a0ed148ad243d5d3851086990691507576b6df","bytes":5177}},"requestSha256":"a9bfa5120bf164024cb7c6052439dc6ff45aaf591415736f88945af6bdff1eef","metricsSha256":"9c3231195dc0b2d717a0571932b9c4c5a69e8060e0036f4b25fc39789b428717","continuationSha256":"c830ef82e3885833714eb32c13d072df04edf7a67a49a847ca0ec84a06fbdcb6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"9ef80dba6ffe0002f5607cf8c7145dc08746b62f34ff4b7c68fa3ba8988ed41e","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"4cb995d0f240877efcea35823f498623dd8a5bd0c8e4bfc34ad0e6469d4dd2ea","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"a9486b29f35d32670ce9148ad946a15c630bbb859dbdce0dec20e61af611d513","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"62572c7f0f9288f3bfc57b95e8228074048c84ceab67e0a6715f36db2c04fccd","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2462423","recipe_identity":"15e90b835282af221334b5852cdccfe393b7a97e4bd9ffc65f46e2dd077b254f","loss":3.344096787293426,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"1d01ff5a4b16806b442ffe976a671ebb9a0225998e3f3fa9e644248893d5c8f1","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"c5b12a7c50326b3278884ed5536251f2c0dcc6c63846baa2904ae1d03536f146","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"4165f26eea32ae2d3a066ee584a771edd6ec188495733165a2dc83bb8344c106","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"4c91c9765d04a5a5705da95b026a807bb2b10865948d4f632915e3f633dc8f77","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"7a3a50e5a2972bffba569fd4d294d32e683aa3d3c304af790eb8f0d41f06ebce","bytes":5118}},"requestSha256":"1b3b891f3e9057ec5ad96c2047fd5a5c67bc8f81322c4db8150f5d20b165691f","metricsSha256":"421eb2d5c97d45696b63733af4a19da87f489ab504b9ebf0a74e33ff4a918769","continuationSha256":"813350c7ae2dc355b3bcaec4852b9a65b7991b8e889b1b2d2a794a42df9a3db9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"4c91c9765d04a5a5705da95b026a807bb2b10865948d4f632915e3f633dc8f77","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"aa30b2b80d2dbc31d8aa29ce09d0eda683fd22c3da3eec3f16a5f17ee9b6325e","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ca6a6cc701ffae059a2069539e5c0da265bf206e1d0a1381622541e6b1fc6f6e","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"15e90b835282af221334b5852cdccfe393b7a97e4bd9ffc65f46e2dd077b254f","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2462424","recipe_identity":"e2931c61a19b774342c3ee6f121993af3d247d8993362aa4911da74117d3ca06","loss":3.3467681063344408,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"236057654ac67ec139b9a17cc3520058772765a6f84e526f56c83c0f54f19104","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"ba393062603d1f16531cf880bb0b1a687a83b122db316d9650a06a0251f4be7f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"f27927cbc317af8f69a074cade28da6bcf30103812f949c3a9ed1964d4db484a","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"010a7e064c5f6fd91f54dc55094a94b32ca5af7a4837f75f3bd59f43da28c82c","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"92fa79c6bf0a83ae7e4a986401218d6f0c54671828227813210dd324cd1fd366","bytes":5204}},"requestSha256":"65157ec721f147d32d54698e60a9ec1150e51b25505ec490d88afc6c89b4717a","metricsSha256":"7b1d47aa7785b5b0111146554e0244d0518f1192f3af4f599aa4c8419ae9991c","continuationSha256":"d98c9ed01eee6f1dc1abfa1ce7ff0658f315ccc50a067f5c1afec21ec06bd1ac","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"010a7e064c5f6fd91f54dc55094a94b32ca5af7a4837f75f3bd59f43da28c82c","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"12660535641d90bd5488e6dd0129db1bb952da247e5454f6881b1f205f52d500","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"268a9a4c6bdf736ca3979d01d5aad949dde85fe8f93efda5995bb980edb9d1e1","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e2931c61a19b774342c3ee6f121993af3d247d8993362aa4911da74117d3ca06","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2462425","recipe_identity":"8363fd14adb7c19c4d2f8cbbf2f80aa2b3737ad0296e6fa0898eb6286aefc393","loss":3.3353408910110067,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"6c834b882cc67b5565e5c4dcb10ddfa3d952f59b76c6d60d3ff00890589a6be0","bytes":744},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"d62c02d9973e69da2eccb90ef57db1294fb8dbb399ed849c889f53bb62595e5a","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"650a758b0ecaec1c936989a28e0a5e29b25d16b8d8fe54482c059fa249fd5a47","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"d36b8040241a797045387b016aaeb74a78909e9c149780c47b23a402d1a6e301","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"99d883cda59176e4b5b445fc9dcbdabb067d23686449a39a8daa50f4a435b365","bytes":5120}},"requestSha256":"2911a52947cadbd36fa7c64e5828771020546732674b909d65fc9990c390bea3","metricsSha256":"119b469c969c136d244fcce1a5e394b181c18226b42807dddfc2f1de11e4ef19","continuationSha256":"288224941fab6e1b3372e227f6836b7aef0d87d0d6e2583415e93a849306bd21","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"d36b8040241a797045387b016aaeb74a78909e9c149780c47b23a402d1a6e301","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"7042e356d7f4c79e135aa021e7a37c82f712e3a958032102d1185e896a4b6b0d","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fe161ab59b1da4324bf8a436c924204a0cf882f05f3305226efc7809a554ea61","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"8363fd14adb7c19c4d2f8cbbf2f80aa2b3737ad0296e6fa0898eb6286aefc393","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2462426","recipe_identity":"e3f075b58362aabf76cc5787bc35f6542096549af65a351fde56847c0e1f9b13","loss":3.329703427837872,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"0328c556108e69e4cc0b015d9a763cce36591e837831788f5cba35ead32a6c06","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"24b818313637b5434e27a5151f478355537e8f7da6595185d5b4594c267a3b74","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"ddb9b81f3cf3563bc4449ccfc86b87c31897d5dccf7318364057ba3c40602e23","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"ddb992b2ea44584854771a31ec8b23f0ae96a4ea433510aa998d6a93b0e9a8d1","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"0a6bb7f50a4e77e1216265c073a8a44f17c149abc17fa1e7a8efd1fd4586b9b4","bytes":5206}},"requestSha256":"9dd25739099ac9b08b51e29caf84928210408b61cd08275f6a7e27b9062c8cf1","metricsSha256":"8d98108c3c10dbca4e04b07b1687208ae84e1722073436b5dbf82120bd1f74cf","continuationSha256":"4b1c5d60593971b06d718f6b4deac9420c56d67584f1460fc4c3bd180c8d7834","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"ddb992b2ea44584854771a31ec8b23f0ae96a4ea433510aa998d6a93b0e9a8d1","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"ec1e818ed10de0bf1b47e696490ffb667579c889535e0a45360d2e8357aee6a3","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d83bd101187d23ee2c530a48d81c9e6473d07a4c3c48783a6129b33fb0b4337d","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e3f075b58362aabf76cc5787bc35f6542096549af65a351fde56847c0e1f9b13","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2462427","recipe_identity":"27aee36d77786b54a3adf62653b5790b61e89f0560bd21d5124e8e4fd7425575","loss":3.333176981799873,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"5064f43d5ed9088d25f091426b9c1c1be892898f45dda0c40d88b9e519d89ef2","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"8c7ccf760f80af11dd03b064e392acb6b3fd92dbb65fca299285dfbec7c4815c","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"ecea9a1c8ae761745e41255e4cddf93170d3af3e1a20479826458ae5e693e5ca","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"533b62c0762a3b54f2fc03cbaf90abdb8ad44801520e7812b17fe37a26fb95a8","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"0925542ec3af65fdbaa711cbe4b812ce1110f481ee94dc430fc90c795ab0cf26","bytes":5206}},"requestSha256":"4e40ccc51a3c829debdb91d327d532036b402b16d1ffbddb18c6709ad0f4f00f","metricsSha256":"4e700b27d8ca727c8e5f5b3040b2f0fca2220b76d46616db759048ca5640d867","continuationSha256":"e7c8cce5d656d6a15d98e981adc5005a57c6f111ef873e58711fb014519abac3","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"533b62c0762a3b54f2fc03cbaf90abdb8ad44801520e7812b17fe37a26fb95a8","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"81a515345f7e36c35443ffbd38e0de33bc719eba40e76df7b6222d40a7196dca","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"247aa962ed21e97713a7560db7217e8936a7bff4ca0f52fedd30c1890b6b47bc","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"27aee36d77786b54a3adf62653b5790b61e89f0560bd21d5124e8e4fd7425575","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2462428","recipe_identity":"9cabb5e606835305ff57e13aac69477f486463416ea7937a4f1c645b287c5e84","loss":3.332162405103496,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"b9ea421268f103b9eab93a025a5b68838339cd450d210c56f8dd6164ab57bb23","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"a2c032d1a316684efd6241c909064f3a66b782335748717e9b3b5ef3991feebc","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"550628b00cb9237a55c96945820cbb845cd5f4d2951e27a019d827c39514c412","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"016d87f62dc132e35e5ba072ee1c56d3f1169fb4f2ee8bb7311e985585eb7f89","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"38edb7ccdf00f230b2274b85904fb7955218b5295ac314fd2d65f99339159dce","bytes":5205}},"requestSha256":"4ba8bedbef75a2e96bdf1f459ec75f82c05d638bf533a63d1654eabd7c7036fe","metricsSha256":"14be8acf4b1dc65d476afe52cc487a3297c666e23ea9031522cb8b0dbc50d6ca","continuationSha256":"65b863f822b0ce480d89b4ddaf081e3ea19ccf11cef5ce3a5cbe420041e3c9dd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"016d87f62dc132e35e5ba072ee1c56d3f1169fb4f2ee8bb7311e985585eb7f89","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"1a561512ef4565857ed8c501f076df0f0be03eb393879abdecf1196a9361a592","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6de3e6db8edada0c1e424c37fde5f1bd4367288f3194346b7c1ea451a9d81654","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9cabb5e606835305ff57e13aac69477f486463416ea7937a4f1c645b287c5e84","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2462429","recipe_identity":"a5cb379de5fd2d960d75b134648b1a33223d7443acb176cd30f3413be60fc7c6","loss":3.3273526546602206,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"add656bd82cc26d4d2d195b14c766766a2843e242ff16fabd17640ad3f473165","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"2b90c5e47274a23ad7b0bc5c54f6ba62a56224135da235ce43090fc2b0d8eecd","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"041cfe2c334a1e957570a8620eb87d5ef2cb5c4939139c309d45a0091a49975e","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"67f2dab5e95eeb8c0b44416ff02eb65520c17e102e93199dab75bbc9ec151d1a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"da784c7f05072b5c150396eced8772dea3f07c500235f39259e2fa307b924f6e","bytes":5205}},"requestSha256":"622fd67439479bb2c69e0d19d5c3d83da3bf5bb711058e1bdab6577641fad8d8","metricsSha256":"211db1f7d061b7bd2bc47875f8c1a62d901a483d71d897380ff1ee0dbcb43b56","continuationSha256":"53bcfcfec98d4feea24a1db4686ae97b08665d8b3c88082185b65934dec75e03","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"67f2dab5e95eeb8c0b44416ff02eb65520c17e102e93199dab75bbc9ec151d1a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"fe387b7c505ac7323cfe178de6446d2fc9f6576f84fddf3cd1bf3cb4de5a5970","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"21a390b5b7040f7ead6db06e2dda44050a69638169b0a62e0db960793e24a1bb","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"a5cb379de5fd2d960d75b134648b1a33223d7443acb176cd30f3413be60fc7c6","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2462430","recipe_identity":"b1103c60c8ccb4fb2e8826de46f48ddd72e7e59bd5b57cc86803915cd89207ba","loss":3.3324995921913207,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"9403bf8b096c5aed63e484b002e794ab5629c2ad406331a94fb384e14ead510b","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"f7df4bbcddf4feec2cd54c1b0788e1ee4f184637822e221356908cf8c150cb86","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"30128432ca03464279fe42d721c330063897b437aefc4abc51aa71bfa022f35c","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"215ec99e934fe9cc342f8e9a695a84a498aa8457c3cdf373d39ec694fdffbe01","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"65cd0fabced3acb7cf0e60b5ffb3da901018bb81cbc23da9ddc44d54d1e938b5","bytes":5120}},"requestSha256":"842762d5e66eef9f8695af68c2c088d8409c0124ad9e8c34fbf7962c559f5e2e","metricsSha256":"522fcdb8cb88ceebda4ee552adab7829c985ce7d8f9617305de9134adb794576","continuationSha256":"993802a2a5c62450fbf2e299eaf8bae4e384b94db8970233e583863efeb01c97","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"215ec99e934fe9cc342f8e9a695a84a498aa8457c3cdf373d39ec694fdffbe01","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"e8daebff11bd520233f319539099311601c71c48e0404251b9010554f5e1a7bf","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"79e7048281e5d51fbff5c98b29e57ce6052e170049fc2d31d6fe609447dbb200","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b1103c60c8ccb4fb2e8826de46f48ddd72e7e59bd5b57cc86803915cd89207ba","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2462431","recipe_identity":"9a9935237f12cf74143188f32d5dce737f4c75b6c907e2a25037497f3f5633ba","loss":3.3288749161129143,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"931d3cbfbf66f09c644c36ae217f9bae8368c5dec83cc99f6086092ae5e42356","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"892ba53125443d63bc3846c062a4af1d1d266245184580befc68b88a7f867384","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"2d33bd5a4d8985d4e2fdf0391690a68376ddb1c0cf4b0c1f4d80b63687f0dafe","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"d502cd29573f5954b4ef69fefe72ebd582d0453a3d40131a6b93d8a8cb6778dd","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"6439573077101ff75d21e1e29b8649bc9c92e5d60655e4a5ecadbea2d50fac48","bytes":5206}},"requestSha256":"1c37976b35895a23f7af59b0facdfca82180e34e9f39e330b4003efd1adf4c73","metricsSha256":"e3938507e13c1e38bf2a24477b4864bc6de55d997b113440558dfd3eb930ed35","continuationSha256":"dcb7510729951d3166aecf26a1fb2c4b102a7eb3c77620256f3f760b8a70dabc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"d502cd29573f5954b4ef69fefe72ebd582d0453a3d40131a6b93d8a8cb6778dd","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"8a925d2d5f46cd6ec46aa29114a3d46c2e37685bf484c8fadb8466508265b525","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5d9de60c4339178f8dba4408fff8face1812b8349e9ee188c7f567cb1772b615","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9a9935237f12cf74143188f32d5dce737f4c75b6c907e2a25037497f3f5633ba","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2462432","recipe_identity":"dcedb75f9a23ef031fcaf5361d3e516a75fc8157aab00b01d6f935878fe61292","loss":3.326803840448947,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"f35708d475e9f54f19a79cdd580852227c59b8f0baacb08e838f799be00ddfff","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"180d4b002f18606352bfed395d1c71c21c173530042dd04573b1f026b15b8c2d","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"c9b8933eee4518ffb2e5effcfd1bc933edabb2a63efd531ce4349d5680316637","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"21b75df7fe9fac49eec82fff29763e055926debe7b72aebd571830a0b81922dc","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"bc399c4568354209faf95480ed5bf0dd5eeb0f52a37e615255c66b18c7a18941","bytes":5120}},"requestSha256":"6ad4c07d49e15466a0c159194f04f70dcc5c33afcaa1a65241178b0a3299710f","metricsSha256":"e85d8a0c4fe82084e20454f7ac75f219adb0d3dfab14adfcc9a769fff445c6f1","continuationSha256":"18496669601fb2fbd097dea9aaa400ed584ede4520c77f4ff2a74b4e894627a9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"21b75df7fe9fac49eec82fff29763e055926debe7b72aebd571830a0b81922dc","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"64cdfd4a2dbaf5b34543d9895e8c7001523e764170f71f2a5599451ffe12c6d5","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1de442cfa5ee313630365e69acc0df9f29d10a1495446caa77d00d874b5dc872","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"dcedb75f9a23ef031fcaf5361d3e516a75fc8157aab00b01d6f935878fe61292","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2462433","recipe_identity":"fb00f4cc8574c8e2e32bc8ac66c5bbceb505db019402f9afc8db9d0ce062d6f8","loss":3.326943668256754,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"81411d0e3f664f83eb879c56c1977452550c1692797da322757d6a8b297ff753","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"3a05da569e7b7ff6d548be1bab2155312d4eef06ec0beeb5063674596b9962e8","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"cc0b812e0b2640080026fb8187fd10066e8971d8a33f67899e375ec095391c6a","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"da9ece2d553a5772ab7810035c5f07a10091b9bcbbc6a2a7ec4ec06964f394e6","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"34c5b19e06d766249349b89bd1c68cbf347b96ccbee53cc09ee187a2b69b1c30","bytes":5119}},"requestSha256":"8fa0de151cb64b7a979cfc5f1226bcd1db7ca778bc55b7b3e371f72fa25b1aa3","metricsSha256":"860e9a9f190b977a5f486dfdaf1d5c80f3a7f3beb54bf5d08de5da25eafb632c","continuationSha256":"bbc48c44e11f0ad9b39a1ba094f7f388d54db5ccd2b4231accffbf4a65b98327","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"da9ece2d553a5772ab7810035c5f07a10091b9bcbbc6a2a7ec4ec06964f394e6","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"20a199a521415240710003fc302a9e4cd93e523c4a431909dca8fa7abc597592","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fb69e174dda123e8dce12f0ed96c3f9773f75894f42825bd9327097db7b0f9e0","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"fb00f4cc8574c8e2e32bc8ac66c5bbceb505db019402f9afc8db9d0ce062d6f8","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2462434","recipe_identity":"1137d89542b5cf38d0abaf20d101d9ac7766b5101f17ef1e798d52dece4e4794","loss":3.3289655426657414,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"ecc464cfbc010ccd07173d95d06f354337ab4f3672943e81564f599a9aa001d5","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"0c30aacf2c43b62d646829427c54cac2a375d22cf26d382b69f6f79885de0fa0","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"c0708ed00f7b4700d1b28197fda9b18dd34842e905de7f76175bb3e5e774623e","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"bc9296d7dedd5eee5fab76a8746f535ea00b27d4852e570f50690fdbbd1a09b2","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"bfcc8a3de8dfae1e31a9e159cdce71b5f5e98b79a185da0ad4e70a7b5ee2ad80","bytes":5118}},"requestSha256":"ae758b87f50a22d878908f3176728063ecdeeef4459e0f7d1319508cffd9c77f","metricsSha256":"c1203896e56088faef1b4ffcca7e519bd530b54c8049c9ee39fdd205abea9aef","continuationSha256":"329c888acecfd9b57a14950d71b2ef354f8a4b553a8d5391eb8b3e47ef54f3ed","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"bc9296d7dedd5eee5fab76a8746f535ea00b27d4852e570f50690fdbbd1a09b2","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"154646a8cc07844d40136970007b6b2200bc23ae9e5f5c8d6166a99b956d9792","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"851ff4781d03ef4df9a3d6e9de4a0af53afd894828008e2909a355f8507f5095","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"1137d89542b5cf38d0abaf20d101d9ac7766b5101f17ef1e798d52dece4e4794","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2462436","recipe_identity":"149d10e143117367645e30751331ff7724787603017158ef92fb3083271da8d5","loss":3.3259936856136956,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"28a9c5b30c6823b4b36bf3859df72f04b42204aabf1e11bf6a8b2439b9dcbc61","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"94187c9e6d96ed8aae7c57d3405670141f8956bf03c1aff705e935bc3268b444","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"a6021c5045e1c56a03bfb125f23fea036932ed56ff01d4d7c5737cedd4dc28e1","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"356734227ce75c6979083e63327a3eb36a0638aaed268034fba92eede0589458","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"f4edc66513969ecb62dca60346438af6c72500a881445c2829ae364568bd199c","bytes":5175}},"requestSha256":"b57418d10fedf62c3235883ff365e1eff66a4e9f06cf5aee0350f5105e25578a","metricsSha256":"da2ea18ab0b819f1ef95c2a2a9345ca2fdcb3a358a2920e0b0316675b99e2493","continuationSha256":"94978e82fa688b9c93b0983aad2f1ac369b0098d2c504cc226f28d40ecfa4be4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"356734227ce75c6979083e63327a3eb36a0638aaed268034fba92eede0589458","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"e9540af513e92de7a69641fa35ffea2973aa932e1e3695790581aa5a9a0e751d","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9324a71daee969af51a0e006c6c18a308ebc49d9d426e66d2b364dc83500f4b8","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"149d10e143117367645e30751331ff7724787603017158ef92fb3083271da8d5","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2462437","recipe_identity":"24c370cb36b7cbbe867bc747cbc3610f16518bc4901ba93169c245ddd29e8fac","loss":3.330497737937856,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"a1b4c00c04e34fb17c7446da89e0cdadc1cc0c5486b66aa9edea80a7cf628c41","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"cae8ca5453305d4c73182a590916c313e13d6182bfa51791ce699c112993b9f6","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"d92bfc548b1a5f42669d13a349e9b97b13c54b584b94bdb902742a1c45ae6ebc","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"8ff7fa2d0c5abb9d794d29344f62dccfe0e5f93e70e5e5aca50261acaff5eb2a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"e9a643555bd49f7658a9e8341c6c9be17275fad15ec3b2970fe3aba6163eba63","bytes":5118}},"requestSha256":"d543302b7e13929eddf7be2e8733f9c772b0c3988c8fb05a511eee07742a3d8d","metricsSha256":"866cbc5179c75d6fc57dc7cda95e0845a4e48fbf3f8fd5efd9717c3ef6ea5f42","continuationSha256":"5dd76f9e8f832738cbf68e5c9f08977819e83e4084045a47b6e459c7fa55214d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"8ff7fa2d0c5abb9d794d29344f62dccfe0e5f93e70e5e5aca50261acaff5eb2a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"effe3647a6c6f526b98a9730018656fbe76cc9e816da4c19a46e76328c85264b","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b927c62af4c3b1d479bc828d23b03c370c13a90ae9d07aefd5743ff07e4a1122","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24c370cb36b7cbbe867bc747cbc3610f16518bc4901ba93169c245ddd29e8fac","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2462438","recipe_identity":"cafe06780dca307d1181ce964cb7e02aa1aebc05afd427ed074eeaea5dc66b3b","loss":3.3256956317066297,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"d2031806d9e8230e09f74c450bd34b0c7fdd4bc360a678b267d8834607a9088d","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"2e5319958b424073cf294c0ca142d4bb8274ed3675d0efd2b25a35db3eef8f3d","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"9d5511ac441c9862e49445cb33fd828c88af123e6b456f2fb28fdeeedb3116a8","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"1181987151d60171196bab1884ee2da68ee2626fa973574e011e7e7e7afe03bd","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"9ff20d44ca6f17b0630706949fab257feb3940b80e5c27d2e8f53f80e87b8fca","bytes":5203}},"requestSha256":"f9328a9236b673381979926a841ee63c6d6514d5e8b28874073847c80c38628b","metricsSha256":"384281e4898b3cf0d405f5c87c17f56cd4bf300484ad78e6d394f8274909a5f7","continuationSha256":"d764bffc7f3c549b54c248534be1cf677a8567f9a4b9ba5b34005b78eb23d570","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"1181987151d60171196bab1884ee2da68ee2626fa973574e011e7e7e7afe03bd","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"4bc2cfd8b0304a795ddd44dd3fc2e06be573a8b57f8d553efd09a2d86992ad38","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"f5bdffede29907151712f3d116494eb8738fbf9aa71abd412c97e6c45712b66c","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"cafe06780dca307d1181ce964cb7e02aa1aebc05afd427ed074eeaea5dc66b3b","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2462439","recipe_identity":"33a93bd6c1968dff1d3e7a6b597c71a09addd1f60bdf51ea217429b55932f96b","loss":3.329253608321027,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"122fdabc707dbc9c88c4dd754aaa43333b01cb786fbd7b9a29973337d76547c2","bytes":746},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"ed2c896c0f8e2e65d32d71d581112fcd9179611be1d3ed1ec0c3bc044ba63b14","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"f6c813e1ab75be437572fff3322cb4492019aa90c70200a0366b2d8cf0cd05bc","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"0fb17df5588e46c139e569bfa4816d8055cead8cd1589fa9da191914fc68e91a","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"f81591648b42be41bb3ccebaac436064476643bf72ccd003c894c845220576ca","bytes":5177}},"requestSha256":"2e4f5d5d225e76b47a4bc0993c707badd6c3b84f75616feaceb4e18fbc3640f5","metricsSha256":"8f3d38a5c7c9bb85a67fef88ae1aaa1b20b03b04168573682c17292793480eb5","continuationSha256":"8b603740561d91c035ed3b132e58899ee99002d07d65237234a2fb4224c6a90e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"0fb17df5588e46c139e569bfa4816d8055cead8cd1589fa9da191914fc68e91a","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"be2f815fcdb49c554db3ea2186f8e05c3651f9eb081ee04045ff2432afd29c77","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"86f77c96222b09e31869ab0b4ecf0dd889917f31bb3a90bbf8a735fbb4a9cd11","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"33a93bd6c1968dff1d3e7a6b597c71a09addd1f60bdf51ea217429b55932f96b","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2462440","recipe_identity":"61fb1e9b63d2b891eeb96611eef0b058c79742cc8ea82afbb3ce7b5f038799b6","loss":3.3312332817107655,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"67909b39444d73d17db74375401cdf7ff843e9e6c0dedecd0c70b013a462bb56","bytes":743},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"65cc3b92a437b6ddbcddb6f9d2c805da9556e69a223d60b6bb1390141914a877","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"db7626a8844256b3b6b522f979cda865621e3e29407743609e4fa555c5e7fb78","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"7e3ee50662ad65c270fda447e1b65e60f2682d1f8b55ab917843ffec74c44709","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"7ba67e94de559c4bd36aa5ad694971055f6a8b4f12ce427667b912bb6f5c05e8","bytes":5206}},"requestSha256":"76ef971d40841403c5c23be4dd0096508894ef1f1752d54251cfb5fb78773b09","metricsSha256":"a953f4ec499d3681152c60afbc21b030c46b585a7b4005ca7ffb7fa257d6fb3d","continuationSha256":"38b9669248a8eaf20ea887918b62b3647b3df3d220eaed889fee260b2f37a8c2","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"7e3ee50662ad65c270fda447e1b65e60f2682d1f8b55ab917843ffec74c44709","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"e5e1b39a095a9d4d7a376937bcb810ef413e80dfb90629c10476ba1a71760457","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"24f65db8a35e4dc278e9bf70200b5f5e9cde065ce3275728f994547d64514669","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"61fb1e9b63d2b891eeb96611eef0b058c79742cc8ea82afbb3ce7b5f038799b6","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2462441","recipe_identity":"7f6447cd47fcc6fe2100ea9a424bff5232d3b049be3dd5d976df77ff195cec2e","loss":3.3268325459296793,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"3ad868d2d1a74b71881a5b697e3754fd6a4b827b6659d47a7fc0b5b0d3fb964d","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"dcc64bc7b850168769e7d0b6318f169602e21c7319c898a3dff01dce106aedd2","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"4e2bdb0ae9260a459befdc72682f32b268038b94045ee8c368e6015841a93d2a","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"5dec0fdb06c400d7aba64c8c8cfa3ad8bdcf4d196261608a108a3e99a6115c27","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"cf35eeaf75e781460cfaa2a961c524b9420edd8cb500cea720cb632bb4ef5a83","bytes":5174}},"requestSha256":"bbc152238f704bf2fa5659aa32f7aea0e52e36f074204661f957306e226abc84","metricsSha256":"d28551db7651e8f30915032a4e77ee9430f7765d6e23ff29a6d0411a4d557fa5","continuationSha256":"1928d4db8fc0d700a032bd205fa9d1debdc0b8915f2f39cc43e2baeda41acac9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"5dec0fdb06c400d7aba64c8c8cfa3ad8bdcf4d196261608a108a3e99a6115c27","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"f966c09c9432bda9059a607abbc6a48c242d15425960c5cdc7d41a3dc992bc71","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5fd83c340a2af6b98372cbfc076970e0e217376c0f8f6b3d658339387ac44fa1","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"7f6447cd47fcc6fe2100ea9a424bff5232d3b049be3dd5d976df77ff195cec2e","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2462442","recipe_identity":"32cdede875b97c0db8c983896499262fe3c0c82ed6b1df23a317dfd586579294","loss":3.330710784572413,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"9c2522b833c56a09347b56d1d36ccb7f650293310a4dda1a1086a7016b36997c","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"a2305f02852c02d70e0e041d967622a197190318d07f0520834860b8661e641f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"06ee3b99e76a4f64a90501bf93af996084f8bb1c8102d4bf1ecc51e39d8f6729","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"98a1211fe6e602db808c1b177b4a29f945af44b9022da9ba3b7da114c5605f27","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"2a272c899f8c090d9fcf0cb2035dfbd0879cc959dac4ca87f0307511b181ae59","bytes":5117}},"requestSha256":"91e50e00e776bad22aec3ca56828453e7f9172642c06b6b337b90e0605bdf6ce","metricsSha256":"56bbbc4e176fe052f0fbb7e5df4bbf87026c858c629707eb3540e6be8e50c2a3","continuationSha256":"f1b5b0fcf395d27e0b28f6f56252a96020871c4aabd24733e007a7b529037c8f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"98a1211fe6e602db808c1b177b4a29f945af44b9022da9ba3b7da114c5605f27","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"9c968917cdc279fda9799651d21e30e78589df23db2d3e96d395a9ad1083262e","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"408ae6b36e36352c63407fa97274e1a8f2fc1146de32a9cbef9c4e1292d30021","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"32cdede875b97c0db8c983896499262fe3c0c82ed6b1df23a317dfd586579294","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2462443","recipe_identity":"6c185218dab7c4192740219d8b2f8c91cd1d7c30a18685b452608b12f2111aca","loss":3.3312423701441385,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"54cd8cca47839eafcef09d5ef58f2b547e1f875eca5d60930fa265f750cf4229","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"217f32ae7b99bfeb80c1c279f888e118169a7fc77e4e51dcd1c61aac3f04f4d7","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"3b05a44c9ed451766dd54dbcad74800625efa9811f8d361cb53f027bff0fdb2e","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"29683d24441865bf2cd3d5c4b4f6db53450bf7100b8a09487bc621ce35dc0ef7","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"e174278ea76bfa879abb22bd6b6e27f79543d7f6c23e5027cdb97a83e7c00906","bytes":5203}},"requestSha256":"b371d21f0c919134333f7a7fed22cd4060606d34f5762183260ec1a753c9bf43","metricsSha256":"d01f973722933f3e3f805e587555fdcbd4f6ea6c29e68dbdbe2c30360f8d4176","continuationSha256":"f5d2a35022c1aefab3c6aa2132ec59bcc668630828249af30a90bcd92fa504d0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"29683d24441865bf2cd3d5c4b4f6db53450bf7100b8a09487bc621ce35dc0ef7","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"fc4226a8f879e70cc6eeb961edceab802b930e964fce1655c7f657357f0cc6de","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"11d66426e6961c59da6f7420074a2c0fca0dd52221ad92773acf0a2f499c4445","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"6c185218dab7c4192740219d8b2f8c91cd1d7c30a18685b452608b12f2111aca","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2462444","recipe_identity":"46a565df52bfc93dbaad3569aebb5a542ecb4a20c3b6710ddf352c53750ae9d6","loss":3.3181250881633693,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"ae8fe13b9baf3207380a6a2bc0c87b9c264334072c557cb9a9ae53aea2f4b91e","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"41065e702e9ef629389c14eea545cb8fb9095ec51b088a8e9355b4d5b5d425ef","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"a3861ebc5cfaa4c2832e427bc67decd0b7d5c387629aa73c6353a8c4982fe4b2","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"0b335161386f977338016f9757990a735b421f344b0da4c16440d89da6e81200","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"f6d5c226d32fdc48db07a17f985f94fd8a477d81ffbf982368a4b01da61711e2","bytes":5175}},"requestSha256":"e30ede5b4ec5f5e1b866d63bc41bd2a5e94127c45c3e45222471f5c4d4beccdd","metricsSha256":"2c573ed8e74994e707a38a672dad0164b758cc58712e11a53218666b6f85406c","continuationSha256":"4eeb18bda3b282f7f82efacbbc8c92725a013c4ee0d2805b2fcc0e7dad7da16f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"0b335161386f977338016f9757990a735b421f344b0da4c16440d89da6e81200","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"a6351decc44a7bcf064dbee17bc1d5e1268cdf5636a7485fbc8f02d82cf5c027","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"0df5f00389d8db3faa01e1d498ad5afb378b3a037ca53246167f332986851a1d","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"46a565df52bfc93dbaad3569aebb5a542ecb4a20c3b6710ddf352c53750ae9d6","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2462445","recipe_identity":"68b1ca5963792cfeda2a3eb945a848a0473c7e08fffd76285fd59abf4b03bc14","loss":3.321727266794994,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"f032f6cf35ba34545002db0809365eac4bb4429c38094b5f6bd2a801c0e834f5","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"decdf834e5306d361a4ff3c06cf9b51a80fcb978346bfd8df05490fd568ba906","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"df471ae42dc74f6c3fedb03ec25190b42e74f7c1624b8fd970a45564f8ec145c","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d46f93b7e8f67646e99fa29b6977b4861bfd34eeb604d1b19afe04fd378c5292","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"780eb7b064a7a536987955f634f2e3762ccae37fb8a349dffce4dfb568932f7b","bytes":5204}},"requestSha256":"ebe7130fecb8d0a78efd6bf54aa561869c38d85cf2e93851d6c817e4c83825f8","metricsSha256":"eabf161e9e922ac58b0e35c9ecffafa618ee0a3f01b8fae75e95826fad93d702","continuationSha256":"105d3b90145ef71f2c1d2a68fbca745c091824396065608be25f8ab2872a1212","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d46f93b7e8f67646e99fa29b6977b4861bfd34eeb604d1b19afe04fd378c5292","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"bce3dc40dae450ecabecb03bd4accbb22ef13a79338e422438fd5b3462b217f2","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"179e6cf55329a778c1fd076e5900699a117963dbd26fbd91da71971f38641593","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"68b1ca5963792cfeda2a3eb945a848a0473c7e08fffd76285fd59abf4b03bc14","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2462446","recipe_identity":"e9a92bef1729a462a0e161f80bc704c4cf92a47dca36ba3271d8b864ec70dc6f","loss":3.3226821783562257,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"b2f9ccb03fe884066ed91f48d125d7773d28f2ab591db6c32805e3cfe01dceca","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"f45f1e31b01bd1c6c11bada1c548c37a7a597f424e22d40798cb7aa05b3a770e","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"9d7dc0b7f49102577743aded9619cee9ea591fbba46c93347436aff076aacde9","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"38febe5c4eb276505b860807cee71c9c556306c07310827784bcdfc7397807f6","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"005ca798b7dd144f07d2998a8e7e160fe6434d1b97ea7b8016ba4b1a11d655ad","bytes":5174}},"requestSha256":"08531897108b40ea6a2a3721b8219503b3f7c656fbc41bdbb536692812a85034","metricsSha256":"8deef4d5aace0ec60f37d0b4656c581e225dce552e8035d33c0d9610031ec742","continuationSha256":"dc02d5da2ada28b4069360bf9ff7e8fec38e52e7d47c379e42f91edd965539cc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"38febe5c4eb276505b860807cee71c9c556306c07310827784bcdfc7397807f6","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"89f364e78d7408a4742d0d9519c58512a937e8d04569dea9bc102c9f58717726","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"bfa888f0f8be12dcebdc78f9997c195f29eeb1f6b8ead2fa0c306fe655aedcf8","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e9a92bef1729a462a0e161f80bc704c4cf92a47dca36ba3271d8b864ec70dc6f","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2462447","recipe_identity":"4ebcd6d09eab37fd01c31c6d19a0b505739af0fd1845c0cdb8c14534be9a8e11","loss":3.3210442227505452,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"9ac7b6be1c7f37a7a25f8a23ff196b0151ea7030b3a0cce44556d78106646630","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"d029b0faed380d07bee7f6485798a58be2f0d4c6addc34cbc85b41b799772ecc","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"554cedc5e3daa03f83a3c4a0b5504f56b85fb41ae4753e29e3b7e9e3b827b422","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"7ca2bd3fd0909ec3dc7ba63a74927a7df1bd1b732670665345dd0a320e192226","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"fec431ef65d871c0cbd75907a6fafdeecda0e73556398ced7fd7767fbe8ffac1","bytes":5203}},"requestSha256":"349e4a93d11d1ce64bc2540fe53fd7cfac3b3160d9ea0b264ce831f6a7a16a7e","metricsSha256":"ab4f393ce51927664d5a7c023a14f58517115011643e26ed0e3532d8a53d7d3a","continuationSha256":"df5220ac9ed2063e0efde47fa5cf487bc26ca8cfc08a6209d78e231d43a4f637","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"7ca2bd3fd0909ec3dc7ba63a74927a7df1bd1b732670665345dd0a320e192226","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"c32a1f29a7f34ddde0b8c215fbdefb5607a00609727a9c4c7484c3488ad1b073","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"ca74eeffa1309f84eae15331922196dfe7300b4dce96c52ae4e7fed76ad97a0e","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"4ebcd6d09eab37fd01c31c6d19a0b505739af0fd1845c0cdb8c14534be9a8e11","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2462448","recipe_identity":"14610ecdb84b64a9afdb08827c7c4c863ec15ed3d7aa44ae58f0afbe9a0ceae5","loss":3.3280224159050658,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"a15edf32d96987d982e8714f73ecfceddaa78f855c8d0d448a505cc217d2d89b","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"9adf22a99b89c8f78bf9ce8f114ecb7de4cecef31ab1ddd54e2362a31dbee320","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"afa4d02d7dc6e2a52fc88418a97f0d833053781e32fd7669edd7e974b78389e0","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"e4c8ee5bae2d0e224fd9a89e7708c71f30f66e7a547a8e978d3faaeab25c0548","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"106870d7383416763826f42dbc258c67ef8034cdc1a3c92bb206ef1de9eaf3d9","bytes":5118}},"requestSha256":"efe9c01e3c77f8844c771491df159b8a72328a4d119fa62e00d54da4d143bfa3","metricsSha256":"5f86c140e165503e5ebe6e587751899ed44eb91ab0e30142b14fbdc6250d543e","continuationSha256":"12a42dc2c23718bff8cf210da07811357635c7795aea1a6f3801ead67e2544ed","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"e4c8ee5bae2d0e224fd9a89e7708c71f30f66e7a547a8e978d3faaeab25c0548","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"b3d0253aecd189346b58ee163def92c8fada2c46a9fc2a4780c4f620715fedfa","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"e34d1a33d6f759d6f390c5b6ef53ee349cdd9f95abbc42a03810a61a7d18ecc6","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"14610ecdb84b64a9afdb08827c7c4c863ec15ed3d7aa44ae58f0afbe9a0ceae5","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2462449","recipe_identity":"08a52e8cae530a555cd1b5fdf675f859f806b116d213f10e13dcda0d32b50681","loss":3.3205936185545912,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"b36404b14896d895a20ffeaaeb1397cfe6a2821009334cf7c2435bc357b6cde6","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"d9bb7d6fc8b243363bacd024889ad553ad0d4d3d9ade958e65479475420963ab","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"b6d206689ba0af9b2b3bd039b8a72a1109491c556a79ebdb579039a1b0f01d5c","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"c34f17d0cc73b029caaec65656a2f2b68a0eef87587d529d260d64e2b5aff220","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"f8692bc58848fed3ce8d95ecd455b52625ea2ae46d6aef136864b6c9b8e8240c","bytes":5175}},"requestSha256":"fbc462b18d5c84d4e84faf478e4bc5fae25d2209c77594b0998a569dc9a60581","metricsSha256":"d6a71ecceeced62ffd40be8f87080be1ac22b17ad0a8be5906db397099179412","continuationSha256":"9dd61689ed27a5dd8f3a014db904ba6045cd409c058911d08ee8aea0d2d78d77","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"c34f17d0cc73b029caaec65656a2f2b68a0eef87587d529d260d64e2b5aff220","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"f9e02ef52d7d08dc10b5403ea038ab44d3304b90ff5216381b84677dbba5f9a7","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d11cd4a00bda5016533ea67044ec83a768d288ef2b0763c834bc0ed5ae0c399b","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"08a52e8cae530a555cd1b5fdf675f859f806b116d213f10e13dcda0d32b50681","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2462450","recipe_identity":"225beeea7659341d73d81bb4d4229810db4fc8a35b038c565581fa3588c0cb25","loss":3.313441292120476,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"fe11491b40ca212440af0368711740b4d5f5451a89f5bcd2a5bc74c36167eac9","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"b7a0e9502cf02ef7104728d603515efc8ebc5b399adc9669f34d7ce132ce10a6","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"5ef69d917f9163a7889d9b58b14a610665f5dc4f9e1b7fb248af43b10fd945a0","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"06ffc82837014e49a40c70a45872ca28abb3b659bb3cc140fa760b7628ec3b34","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"e53da46c9857478102f626280b2d8cebc856b0671b2b55c952b9196f2f160424","bytes":5120}},"requestSha256":"a87e2d92efde34e151678beabe7a9a87fa71bf09a19fa49e670fc41d6326c07a","metricsSha256":"9659504bbfa194dc202898e5ff14ca6566488d541a75a2c037dc6422b14f0d3a","continuationSha256":"23fdbbc53c92ab412841b642c37be6e09f63ddf4defd731ad4de34cc7450e487","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"06ffc82837014e49a40c70a45872ca28abb3b659bb3cc140fa760b7628ec3b34","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"01b9afb2aaaccb32b91605a5219002af00bf2a05be947ed4dad82d3ca71b169f","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"b7c465db5c2e8407b878538541c8b36ac874bd9de262b682d4a2bc1376a6834a","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"225beeea7659341d73d81bb4d4229810db4fc8a35b038c565581fa3588c0cb25","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2462451","recipe_identity":"30ce54bde1cb6364aae7467495694e02ce5ecb77f344ef87d9101bc651d7e9e9","loss":3.3209763697506856,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"a4258424dc60ff72459f8add7c7e50c808a0d167c65938be25d6a03d044b442e","bytes":749},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"9e615f8ecd9efba05acc21b95110ae3f9003687da2dba7a45ba85a9ee6c223bb","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"3d4364d798857e591377ac39a7fa0e9d2ad48ec1126347de065f04f3facaa678","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d711c1838ca9a63423f63070483f1cf7dca443e386ad7a6848e9bdefb856bd66","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"f5c8ccb9d1fa3298d9069fc7abb24ee07a6aaa41de532e6fbd2d38b52ff18505","bytes":5113}},"requestSha256":"9473c73b0d3a787057937e893f31f8d1e4b3052af03109f551f0dcc82f765dae","metricsSha256":"7fb30f4277bb8eeef186aa781041bc5ac105e47a44633dba78ccdbb701bc5bd4","continuationSha256":"61f11c5bc4d7199ea2599a25dfe935de79f5d8b35002cdbf0c16fdcfb192f430","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d711c1838ca9a63423f63070483f1cf7dca443e386ad7a6848e9bdefb856bd66","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"fadc692add808af76a75a4b3d1d39ad25c80c57839bb6ac7222ec2bfd16a7662","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"5ae69c1395311365e60244b94ec1edad2f1d9e1795234f0cf2478b8af5dcc180","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"30ce54bde1cb6364aae7467495694e02ce5ecb77f344ef87d9101bc651d7e9e9","loader.blocks":3188096}}},"sync-wsd-h160--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2462452","recipe_identity":"91e18b54ccdb718e27de172c3f8177b9bd4a87cddf785bbcb2ac9145cad5f8c0","loss":3.316854902655363,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"ce18fb819f0de99354ceb7c7325b43b99975c506ab1c1a2353867ba77fd49c9b","bytes":747},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"e02168a1accf11280668bd60215146f4c5ff5636ec270f826b03b3eb178af31f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"c79d61a2be204aa1cb6996b84b61b802a9730de6dbdf93ba24a3e12f37e1a333","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"1968e2b0961565a8dfd4bfc265e36b9a3bcea8636496b97a8f7d0030415dc731","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"1e1c655ee2f49a4b499a636e635575025bcdabc864a6d805195e817397a875ee","bytes":5199}},"requestSha256":"de58af177ec3117aedbb75bb5e0b988800bd34e5315bd97f64ba99e5de89ca53","metricsSha256":"33a1f1cda3a1aa97349dc562341f4c98b0e4dc9b338bd10095bb7641bb6a7564","continuationSha256":"e59a75f8244562f3e985c36fa7ad46d4830816a05eca7d3bf8b7a39ef0c5a9f8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"1968e2b0961565a8dfd4bfc265e36b9a3bcea8636496b97a8f7d0030415dc731","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"754f060595e996be2462adb36b634204d4be29fb5f9fc6b96e71201c8e1e0a77","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"84c3092ed4e4a0c32ec5f2f1b7c260bfc60da483835f5dca951fc5d86fed2299","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"91e18b54ccdb718e27de172c3f8177b9bd4a87cddf785bbcb2ac9145cad5f8c0","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2462453","recipe_identity":"9b8a20615b97ef8dfbb3113a089981b1848b7724e5314880ddfe3f571d8ebc93","loss":3.3143719220945855,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"ddd80fce6ab24ecff34f0676f8a0200433bde5329559865359c99d0c4f933f1a","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"a76fc3c13bacc25df020b71000ec787b284bf7fc187e2a6128b2c3f705fccdf9","bytes":244950123},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"ab9caefafa9e71b1edcc995c287de088ec9c488848aa232eaaece5720fe873f8","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"3c4103b98707b20acc1ad22d3ff51afcef6c8bda293f74ef1369c5550aad60eb","bytes":244950123},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"c66ee3d412e1d3fc3f541aeff6d13b6b20a9330cdcf7ab8d0c3075b3efbd335a","bytes":5199}},"requestSha256":"7b68f5bac36ad65fe8e839ef261b4a119a91a146964b6dbf5cb94670fed59321","metricsSha256":"5997c90a10d9bd1c93273db9dba37298c18121e2e272a869677a0600ab5fab41","continuationSha256":"79ad15f77485e0ea377148c29a5d0af47ed2c1a6d70a4e9786efc8ef74b3bba5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"3c4103b98707b20acc1ad22d3ff51afcef6c8bda293f74ef1369c5550aad60eb","bytes":244950123},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"be59ac980b1a849ae2ab9aaac9c694d9467aab18106080c6febd2ca209a48cad","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"d2f482a3e2d0182a43255ef0a86e2682d60bfe98a21ec1b4c32d4d7ca8c2bdb8","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9b8a20615b97ef8dfbb3113a089981b1848b7724e5314880ddfe3f571d8ebc93","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2462454","recipe_identity":"9c1a4f521cb7918eb2447254a63dbd16f1afb1ebe86de2a5335c7e1039030eba","loss":3.32156990315758,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"1fbd68692c78b1da7185b9cb45d434536a1f50fff655e1bb671c97c60e0f0060","bytes":745},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"adb9c7323e8cc082e5a8cccf49225de1feb070561c0caac91714ad674589c46f","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"8accafee8551c5dc94e191aa09b15aa63a68a8150808ef273ab265fa48f92f9f","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"7d5b87ab5f6e4b63e9593dcf9c97f760bb3518b16a427adffb3d10c5d1dae44f","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"97768ecd00c498199afcd0e16ae2f94f90367ebb20e47616e2e5db54e0738985","bytes":5199}},"requestSha256":"f9b89afe1f5ef01542f1816e6b09ac8e69af3635f16017f02296b802c885e7fd","metricsSha256":"ad38dcfa4bf8ba85ea4891add1f0a5bf33f972031dfa0ba417ed1467aa0d1e57","continuationSha256":"fd2f99c9d8d9e22c95a00b22e14d91fa6f89a7b54862ea37b0642eb4c66773f1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"7d5b87ab5f6e4b63e9593dcf9c97f760bb3518b16a427adffb3d10c5d1dae44f","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"d18ba811313b33f8c05abaf83f4f4076ce1a554fdc514671470bab4c8c9a29e7","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"42a550ef7dc3bf3186f508f9727fb9647c7924ad44e936583beae470cf9e964e","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"9c1a4f521cb7918eb2447254a63dbd16f1afb1ebe86de2a5335c7e1039030eba","loader.blocks":3188096}}},"sync-wsd-h160--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"5cc9d7b3ca0eea6020fc13cf4cf2095cc9a726c586897323324d4f37f7f65a70","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2462455","recipe_identity":"537aaa0750c752af25d395b08ee641c5a03d116f79f0fa8da77cb228737ab36b","loss":3.3156719160106447,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"646000c5ca84578004e466800df659668ef1661cc4e34f2cd132bcb92e4dc629","bytes":748},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"9e842e1e3509100e60d7704941ae2ccf015bfd62a5b8ab10800dce2fb28e8807","bytes":244950059},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"43183829d6c8374bc7dbfc25abc9170ce719961cf2038ab123a53ba408a6bfc5","bytes":81621488},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"6f676a812712c549cd42dcd5f8ea2d2d71eeb5526127823faeb0c2e8f15f71bb","bytes":244950059},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"b575db0e75cc2ba1cbdac0de0cf8e996c407e1ec39ecc001298f8806c74682cf","bytes":5201}},"requestSha256":"5901fcef0cb33348c833fbdbeab101211f2b0284c29154a0b5f75f07983d2abb","metricsSha256":"ac7fb77852ac9cebe62b5c9b9d59de0b5cec4d41497de4ede82809f9274f98a1","continuationSha256":"bb241f2231e60471c860103436ee732318b295240c8a9fe5436f6f20d53a95fa","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"6f676a812712c549cd42dcd5f8ea2d2d71eeb5526127823faeb0c2e8f15f71bb","bytes":244950059},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"7a10f49722faeed6a65ababe28fa211c8b84537195b7479a7a42b7ac82525006","bytes":244946027},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"03a0deac21eb5680d309131f90d4177fa7c69de3bab1a59b0d02e8c2976e024e","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":128,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/sync-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":null},"recipe_identity":"537aaa0750c752af25d395b08ee641c5a03d116f79f0fa8da77cb228737ab36b","loader.blocks":3188096}}},"awc8-wsd-h20--0.01--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.01-s42","horizon":20,"attempt":1,"job_id":"2478551","recipe_identity":"2208f6080e8c42a6ad8ca966d04070c3b7ec4b832b14e8bc52a04b9d7821f419","loss":3.7399308415913497,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/result.json","sha256":"3790026d5b9af18285b3b3108dc52b5d54bbca83022e0bd3efe3740042b8845b","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/final.pt","sha256":"139cb46cd3e0473f8784e003441ad9d28186eee762c838b7bdb88b3a1b2ec661","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/final.safetensors","sha256":"18b3c85ec09dd1a71f589d107142220844f2b0b63b08d8b14f0259f47d1d2f52","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-000/final.safetensors","sha256":"6c390d6e83260cf31e98c0440b7bfe5e6dc881989db64491f412820166185c4b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-001/final.safetensors","sha256":"59c248d350357c3784810f03ca1cb03449d2fef3cdcc64464f241afc98fa31e6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-002/final.safetensors","sha256":"bfc766b9cd5ea05955695c0804b8b02bb347161ed7347b3d8cf3b8b6be46818d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-003/final.safetensors","sha256":"1c74418db934388919756057f0eec7e564ae14cd635d3857da2bfabc1c6f3b68","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-004/final.safetensors","sha256":"1d3549cbd125de5a6a2c7a5f2e3dfb00b12c0bdf52361a072ddb802a40a756b6","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-005/final.safetensors","sha256":"b7445bb3bf9840d018feb35c14166e64485c129e02afd4ab7bbec544c6984012","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-006/final.safetensors","sha256":"0911981e093cc8dc625e5e94f7c8d3d0932a4b7f52d355672e8d3a4dddaaffef","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-007/final.safetensors","sha256":"3895a3edd453dedbb99e35f534e86ce15f2a6f0fa543099b2656fcacc397fdfa","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/epoch-035.pt","sha256":"bd230863c153b48abf8ea0d7c2cbd78f6dccf06a2ca8950a1935c4d75f6e56c9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","sha256":"f6b3859ec87b0d14fd812585f2d19ba09b50b5aa3323224450f67da580fe79c3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","sha256":"05e345de8b6033906b7fbcda44e14bd393893b3b62f7a85c59515f8a7146b202","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","sha256":"d83d49302d252d4a84d1aec296d740397bb40022123dca8a4adde314b31b41db","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","sha256":"ba92b05852eecca7dfe117605575b4c37ae3de61a80785b78014ed278c71d4a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","sha256":"afaad04a84bc47b63cf7cf9797592a62a445f538a2ee4acfcd4e442349737374","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","sha256":"18850398fadf397961675e8769b22c887349e80bc0fc687a7d334548058bc19d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","sha256":"ed1a718628052cbe3a4495c452ae5bf3a3e35028bab952c1cb71112a93bdef21","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","sha256":"5d3938e95ec34f76304fdcd1a6b5c5935ad3651481e4e7d9c57904202a0efc8c","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s42/attempt-001/environment.json","sha256":"40c7e2a4b0685083534c9d868ce48a3307967962e7f6543a9f42962ff2546bcc","bytes":5312}},"requestSha256":"f270e00ac5bbae682abbaf5a7ce3aa84f4508ed4b61583e90767a0fe1d260035","metricsSha256":"3e0b2c7dc9b5a7e2f5b16d5a3c7c0f7970d5a4a5bf64c100ec14b8d31893559c","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.01-s43","horizon":20,"attempt":1,"job_id":"2478552","recipe_identity":"f86bfca3c81979e32900237ab02bcf26c30877fcbabde43f67651ddb704197f1","loss":3.7291007105153855,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/result.json","sha256":"1fa35a60de02732aa3b574f8599576dae443f86487e4ba4be2d1e2fad443e5a8","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/final.pt","sha256":"53c35f718f3904ddc431a9b4f96f225a70782b608d3a23e32d1e0ba68cbbf1fb","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/final.safetensors","sha256":"7cc1f1bf26e639893c5cadd9e1ccf5cff1104c6b57eaf7e2d24525afc4ffd958","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-000/final.safetensors","sha256":"fb3a35a2644cda885276e2479dbcefaea38fa769253eecaffbd5446396112107","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-001/final.safetensors","sha256":"e4a21705940d5ce12c5e74d04aae5e649f76b1ebff253083e4510e9de373c75a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-002/final.safetensors","sha256":"67ceee68025b8d1fdb3e001285c2c95893b73a720b8710b9a18570e9a0375063","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-003/final.safetensors","sha256":"22e78e8abc53486775999dc0ab414fd6d0dc736da1b0673b47904cc825b49597","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-004/final.safetensors","sha256":"873cb73896d04297be808afe1b684900354a60c27de56b518738115b28d08bdd","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-005/final.safetensors","sha256":"a8601037ea82aaf189ff8eaafddcab7f038884d18aeb16a778423baa62760195","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-006/final.safetensors","sha256":"5d1e39631933c8ea6f47e6b2d8b8a52a88dd55b43210cd2cbccf83492115542d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-007/final.safetensors","sha256":"713316c0bb95d1b2fea74573d9355251190670ba1ea766c64ea1836b177f9c60","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/epoch-035.pt","sha256":"68035c22821817336011baa4f1c9ac34719e8487e6710321c9b85bfbf4999922","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","sha256":"74e478d5046f6a1c02a55c6e3d8ed0eeff051191924e1f2d91f66a773e793ab8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","sha256":"91f84129ee15e3516c5729089deb44600a8733ae7afaa75cb75537ce34c817e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","sha256":"e90a6fec07a2d2b2181c9817140ae0e2f5d0aae2ad25fe6475b777fd0c8a9784","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","sha256":"25f7e17e0cf787d5e5506903f557420dfd60b1de2fbaf4101959e670926b6b3d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","sha256":"b9927a2837dd0b7d64ecd543670ff04df813e461eaa75ee1898a5ac1acf48223","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","sha256":"ab2fad5cda0a6139902bb7b99207f361601e7bd99cebc9842853ede5eb6be226","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","sha256":"c218c994fdd7df193a0ca5e27dcc1ba5e295ea7b07bb64d7de565f3de4c5133f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","sha256":"fb3f74132f1840933c0d04ed63d5c0f85c0045821d299aa942abb1f224525f9b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s43/attempt-001/environment.json","sha256":"74195966095a8cc49afca554949856f7fb79787cabe42076b6bfa99fc143e169","bytes":5341}},"requestSha256":"ce5edcd7253683e97239d9e640061fbd80679ef73e341eb9a84d5dde1aef397a","metricsSha256":"c067ae8273449ea1177e72ff53694dee0ce534bab7f3b5283a60652a521a9ade","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.01-s44","horizon":20,"attempt":1,"job_id":"2478553","recipe_identity":"55c81c1fc0c3354bc198bd8be1c7a769d09950befcf3fb056ac9f6f254d2e587","loss":3.7392321073116714,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/result.json","sha256":"189a4c304537081479e387812c5bf67f174967a8f40ea8c35bdf6920cf3482a5","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/final.pt","sha256":"4390d0be16ddc4f2b06b3be8d3067b0f2a2d24988b715efd7d1bcb0719388bc7","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/final.safetensors","sha256":"a9d8c538bc7f682bafdf9cefb4614ba4baa9041bd20a2f6e75541261f3d7bf95","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-000/final.safetensors","sha256":"315e153e6497f0eee30a5a03a15044eece82da94fc46b0c9a9ac3676a7115935","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-001/final.safetensors","sha256":"b1fd08d3b01339393d9c4d3fc1e1b49abf9648cde8fbb6052c8468ecc29906ed","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-002/final.safetensors","sha256":"1bc1040a5f19cb373328bce9a10e4f0e433d67626b969a675eecb2e7b75eb755","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-003/final.safetensors","sha256":"fecce144ef37ce994decb54637e3b6e0e1c4002d3c58e567796533e76e12d30c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-004/final.safetensors","sha256":"5314064825dcd0848e4786e39782c6113a7ecea7cfa01a8a75da24fe8a9b2081","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-005/final.safetensors","sha256":"05d347c78fe84e09daf59156b833a8a38dc727dba383a118c22aa824ef414617","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-006/final.safetensors","sha256":"3b98abeabf8e7635a08ba47ca8885dbc8bb778b47ef14db2f38d636faa257482","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-007/final.safetensors","sha256":"f68d150dd9111543fbfb40789b9a9b7b966493e8ddcb95a9f683b0034c8d4495","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/epoch-035.pt","sha256":"b077d44ae3c5a4a38b78be8d78ad0d85a547555c1651c74299defd537772e8b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","sha256":"10af5e3a0c655e177570446f737f39227de6009c6fe34844661cd4c5800c377a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","sha256":"f76e1d3519347a5d2b6c731ef99581d4293a3a0c69804cfa65f681c7262732ef","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","sha256":"8853d8c36a77321a9555d18f65d2918bb6c0062a68e5a4eeac0e91411d70cd4a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","sha256":"8dd8a7886550626c49f5cc58ea23553cec28bffcc46975c2b177c8d6681a01a9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","sha256":"f3181713e4bdbe1c9c488ab0f216a17003566d9ce5eba98e1f4c6c6259915320","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","sha256":"63f6c9a820f6049d9503a856b9c6029770efa2dcae5b9052b1bb272357f435dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","sha256":"1180b94fb3c66484962aa2acac81cbc63ddeb700db65150cbf1665e2d0ee5e88","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","sha256":"8d0565170feae885482149e2ae7bc5eefd17d434b75ff2925db6ed9b293a3239","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.01-s44/attempt-001/environment.json","sha256":"5fde5ae52d2fa005e3b73255e97eaa2cf2618939eca08ca240b4955285842795","bytes":5341}},"requestSha256":"312cd729a1722326bb7badeaffc1e96a4db4b5bbc7a156984be6c42d0cc172ee","metricsSha256":"823e0db9e8e082280477fe8d33792d3f39da26f39a9bc992f0bd35b80c8d48f4","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2478554","recipe_identity":"ee478d981f22d0edb82fe86f1d81a81d5d3d964d20e88b600d237fc6f8997f69","loss":3.7272634253307966,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/result.json","sha256":"7b7bc71fa36019694d0dbb2351a0a9a4437122cb2db23554d271f8424312d5e0","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/final.pt","sha256":"89257dfcb26c790a6c0450a0e77c3e82278a385f1acff566ce5f2f3dcc9816c2","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/final.safetensors","sha256":"65dc42b097883490c233984d3ed9b62f63d276c0935fdc1d89216d55665927f4","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-000/final.safetensors","sha256":"484746032f7e9e780d8a6ebda555244b07da78ca77067def5a59eb5baf237592","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-001/final.safetensors","sha256":"cb61428f1c328d67d69f22f7e9cfa4a72e0f507f3dfc8c10d737d8e04eda606e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-002/final.safetensors","sha256":"e340157690a80598d8998f72cadcac48616d6e3d1ccbbe985a1dac8690bbbb9a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-003/final.safetensors","sha256":"88fe6ac9fae8f46c9919b984ef9c2a6dd12a2739d2f845b9a6bec8b9e6fa879f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-004/final.safetensors","sha256":"fdb75521a8c868904f026632c55a06ddaa72ee671a8015d598f4c785c79872d6","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-005/final.safetensors","sha256":"60768566cae3b9a8ac3e095b29e70928c0d87d9d367b75706cf907aadb640aaf","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-006/final.safetensors","sha256":"5bae9f81eb9f0a9f125978733e113519ac0612798c0f1e863a0d9f7d84883d38","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-007/final.safetensors","sha256":"31aaac14543e2dc75a7a7467d90fd89fa8a7728ea0f2822850d209de50a54a07","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"196e364c428500961ce748437eaf6f33dc54ce494dfbba0b590cc6c740c81aa0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","sha256":"5e1d73ffcf99dbfbc44d3b90dcd9a62734d337508591834a4bba0c162df66c02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","sha256":"5e71e816f0cda350d90bf27ad631af08c6900c9a3867ed048181d8682c8d98bc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","sha256":"3caef4aba0847b6bb39aaccccc94517f4ece845074a013770e04a9e0ebcf3433","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","sha256":"85ceb3cc4f124f5e3af477522f61a8984cedf06e79cac421707ad6d24c745956","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","sha256":"47a5b1573dfb62a639b68c42ee043c4b7a6306e6f19df0d7794d2a61983298cb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","sha256":"489af461b893b78ab574155ae73878e7b5c581d968785735bc80a27b3c5272dc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","sha256":"2e50975e01fa4345818551031cc4e7a1ac367650f1fa33d524e195440e5f1d47","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","sha256":"726f5449c8e6c0065a6c1985f313b6e9230eb039d4fff2b4b1e7b02a629441a2","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s42/attempt-001/environment.json","sha256":"94d548a9e593a1db6de889bde00a6ad024dd674f7949ac623c430841cbf1c8c7","bytes":5310}},"requestSha256":"7943c79d620517596f8a00d39a3333817b2dd809effb6ec1f397434521f43c59","metricsSha256":"b67c0ebcad754fcfeb1db05004acac87bc7eaa616552ee74a2dbc94fcd84c47e","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2478555","recipe_identity":"6dcd3c70e5864e86822834e44eef6ed90375133c38e8625a3ec96627566f50c1","loss":3.731839734352052,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/result.json","sha256":"0d88aaf973edf22784edc83fbee9a75a209558ee676f72df317180c02ce22889","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/final.pt","sha256":"7d3068f1809b105f4b5bacf435c7547c3a321ce308b1f388a067d291490d9972","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/final.safetensors","sha256":"9edddfa8fc513b1c3c26985dbafd78bef9e416ba69d8fd20fe028811ee03d17e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-000/final.safetensors","sha256":"66cf1c60bd05aaeb199ed2e71417dc1938cc5975cb1f9565077cca4715350d77","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-001/final.safetensors","sha256":"7f719dbd9b6ab2181d519a82ece6085045bde7464772ed9918663f74d06113df","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-002/final.safetensors","sha256":"db7ce2d46bc0997425686d8e63a780b8968f24a622e093d16491f217b7b18e68","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-003/final.safetensors","sha256":"615d3d8a5721851eb73c8b1633e7bb113c4ec4f82bfd31edcba44a5512929549","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-004/final.safetensors","sha256":"30c58189216812ed3a7d43525c15f6879f99e1c9a4b61aec18599c2e64ba1008","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-005/final.safetensors","sha256":"c4e21ac2147c1e7fb0aecff8e2bee88f794288303d3408209ce718b62f1ecde2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-006/final.safetensors","sha256":"57e8fae6e6a587d78049b5b45bf0929993a04695c4b9d46a8d8d6563997172ea","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-007/final.safetensors","sha256":"3fb328306a3999ccb2939e47f36b0c588b1874b4bed6ad1d64f4b8fea6cd27c8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"1a7d32d46de9cbb25a7904376456665a87ec9cc81cc1a06fd1425e249d1c2d20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","sha256":"4a07c4ea7b3c25183d79c058a7ce3aad3024039d715bf8bfb7863b69914986e4","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","sha256":"7d6e24c42067d2f0b6a9184f68be5bdee7cc9c530647fde3a597bb8462cc6274","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","sha256":"4cece7531740a5e9564053a120d9d5ba6da9521bf4ff6db63d5fce903bd9a3c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","sha256":"7837f5355f2f1aeef0df5ab202b7908bfaa68a42ea9093083857fb329685dbec","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","sha256":"ae17112faccc7c77fa66e0680fa8c224c55df34a7e06f8f45ff4a8944288e5e6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","sha256":"1bf3d3c3b3177381a2e8955a5f844b128790e60ddfd81853e6dfc04ae906795e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","sha256":"a435c060f737ab4011d2f624ec06eb81f39b3a3148d280e653f51406579576dc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","sha256":"f308d3e4a738009386c96007e6f1c299fc7600083de9f052745125dbcc17d91a","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s43/attempt-001/environment.json","sha256":"1e3c0ce99bcaf728d65d64ed85f1a33e6dc68c443276dc7842d16435c5826cc7","bytes":5255}},"requestSha256":"24b5d27967815c0f176bc4300127106e820867ed310035282636a5537681d33c","metricsSha256":"3e1d71d9bec878a2bc43307ac60cbcae042e755d9f237a77733a0309af1e195a","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2478556","recipe_identity":"007ad94a8b0d6d958780fa201a5c779af61ad0e8f116860cc6d2401f896dce5e","loss":3.7365845429544415,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/result.json","sha256":"dad33631b20d0f5f92181c0a12d4117df1ab55dddad0c75902d8b72c18c52c02","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/final.pt","sha256":"2026ee64f8345ce1daa3a21ae582de213c00fdabfb39cde17be8fa52a447b084","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/final.safetensors","sha256":"d93edb5ae7224caceb4c3bb5f3007e153ffd3fd97d4605e71b69f977436c92c4","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-000/final.safetensors","sha256":"5566ab82474bc2f45a32a052aed701a49983e54f8a6a6dae28406935cc495bc7","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-001/final.safetensors","sha256":"0c0245ffb63d7bf33a67a458381bb7807bc46dc1cf7ed69ddd01eeeff71c632d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-002/final.safetensors","sha256":"9793795ba57fbb56cd6b7c99587e005808cb97dbbbd0410d949c8d446ff4071a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-003/final.safetensors","sha256":"ee4ab2009f654919838115c9df2bdb73f2fcee889433e80ac0724286b8d95ff1","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-004/final.safetensors","sha256":"628a0c3227c767fb8d6ae34c80eb0f0c1a36de4f2524074f70c13382e31aedc2","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-005/final.safetensors","sha256":"3cabbb789d3f4931d1f459c75e8588593efd62d2ed2d8147fb64060f1f74ca6d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-006/final.safetensors","sha256":"39667393a96ac18435881c67fa15de55693c49e7cccfcd84288d409a6d8d3fa2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-007/final.safetensors","sha256":"7eadeb95fd3c50876fe2674c6515894ba3fecc7013cc62a016357a5f85e17406","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"a2ffbce6c06591a737a4e44b2aadfc266727ce3912695c56bc160814b9895b72","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","sha256":"6e7d1c3c36fb40855fc7843acd1ac3d91a797dc57a4b9c280e2ef39bd87ecf0b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","sha256":"4c1e10b0b75bc74f3f15a6d675f8fd2e9766c12e1bc71098a54a90b51c6f5749","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","sha256":"6be6c289573c76cdb8e73d0b0cdbba927046e5a0790a4a537620b5352354ddd4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","sha256":"b7e8d7eebfee0bd4b7461731b46aba58970fa3883753477f299dde44100190e5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","sha256":"7f7efa1b7ed895d804c8efc5dfff957b766e65170fa18889ff6ca35af1998f80","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","sha256":"45bd82343aa0d21611f2de4d506c91e48f3f24b947ab8bc46b2773e1499a88c0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","sha256":"38d5e51c715a51bddff37f0fdb8b911a76edc07eef35b4f3d13b877ffa592984","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","sha256":"4fbc2cfa63462cf8b89979032062ea7014d76691bdd6549391ad0a8a4036e77a","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.008-s44/attempt-001/environment.json","sha256":"1421f6751ffa8988dc3ba540dc160ad709fd61378df5611456b9475e172b31e2","bytes":5341}},"requestSha256":"807ee43d426098aa5cbd0ea287b5318dfdd47ecd46299e80c389500a7865df86","metricsSha256":"0b785b9a7eee8a0413714abb74955212383ab208014a4583f9abf3cea19c6fed","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2478557","recipe_identity":"fbe6baad5480f18d5b18edd60930c22d58698fe4e567d2df10160d9e4e57708a","loss":3.7211869087289102,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"d27f33661998bdf9633a4c12f7d3e1a0694a1c97051b95dfa1b73f3a5e2fb9a1","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"98da10a4c9e102fd9278f0f86fdc51747b48965bac77bfbae630a63b1a7d4247","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"a379688f5c09a612284afc5379765126aa5bb2abe10298102240f8c26f5c2537","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"5a99666f32582d478283078c24aee6cca6b54d12a43f700114c6a3b8ba7bfebc","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"8059533f395db326833ebac8896dd7adb2d9203ad99ed3b9142e6129054c1bcd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"e3f4409ff65398169f85cb11a5182893ed7c1de613608ad790f0d2d6f99cff46","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"0f805ae57167c9309d6fd100a146a4710aeb7e719008735a5898efbfd040d99d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"83ecf3808d2ce18cfb4de9c9f02363512c2d8e50c8d6088f1f43daf8ab45c80c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"6c3043cab36379bdbec152fb923aac07cb5c1f501ee878434f4d17f4cbc0bc5e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"c4acb433fdb01f02c36b0794c253cc6849b9d95c1effe4bf8e0d7adcc8f58a52","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"3837e7df4659bef8d51daa6e994c504dd9e3e6d10f5e74729d2a5f978a22375f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"7497b9b5a572395ad291d1bcdd5441bc3cd0487f2ef97de461803dac274e9173","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"f53c16354670adf826a65353c9d624d8fc35e8aa7d019a89d40ce3a2446dddd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"3a4183347a2dd3af7fa3da53cbee917d6ab5372fe654cb8c92f4e94ce03ff1d5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"423345be27b8fbbca500e336f476fc637400607d63742094827e05b49ae943f1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"c5feb1b40580bf3e4b9d510824e7a30b5546270e187a741b37b44dbcfdd2cea8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"4985b79b803edb24d37544eaa8b75e4f0b4df49e3e4350e9a1d0b6e70fbb4513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"19358ed29d3ed5aae267a685e2b9cfcc5f44bac909c6139ae40819491b8632a8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"3e5a5145797ec7c792c43ca493559e00ce2a9e5e8f46edb1068fb5bbfa451125","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"f05bbe00615c46dad0858094d7904a3e3c44b3fbb9bbf4583a7c7909e3264065","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/environment.json","sha256":"d0fc6d589cfcaa97e4bbd31076a88082ab10111871c0d120d028ea3bcd65d6de","bytes":5340}},"requestSha256":"09b5a0bd6fcacc1fe20f14d16fe392bc09bb1c9ff250e347d968de94a93f6f47","metricsSha256":"ab14d622553dbdf9370e624c26d5ae1d25575a84a90d54b1484c76af61be02a9","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2478558","recipe_identity":"b4c77120ba532ea4f6cf8c3483edab4f9ebffa4d896debcb92476ab831d0415e","loss":3.7336587401190826,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"6e4c75925a59258cd4622ac5333609e4324d7ddfaf771240949d158e871e491d","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"6a3f389708f0cbef4c0be554f61f6663049fd1f9db30eeabfee4c6c325a3d260","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"2833be3548e691f7c4fc590b088d1fbe2e6acc318539f4d17e0ed3df09195a65","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"be1ad64c8691456f8b7cde35a894667310496772ad0b69aff3e52aff102023a5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"0654f73b81f4361ae96ccd212f9d97deb55caf68a711fc3a0f5d895bd296551f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"2e3f5e4eb164c89e95b314bce371656c5666da7a6c6e567413072bc6463b26bd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"7bdebf1cf219daba1fc073fc9e092cdb81d384159df830116dbc3799c76bd2ee","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"2892a7e48e1ff2a490b3d94389a9adfc1fcff536ce00a43bdda2041c030f412f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"f557cfcca61fd528a9a6e26e72cd8a9ee0131a4a9f8a3d186ba09cade2cf736d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"24aade56e9ae9e03d0d98d55c1bc04c24acb3e59e81b4029887869e55b957603","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"0035394882ee3401971bd380b5bf89c22299a53b319e4178dc183fed316e1f86","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"75be3461f98df19dd062f7603c0d0fddb64d238326a063e04658df567b34fd6d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"ee8de29472447441c4fbfcbef0d27c0d8898226097d5dfe2f9bd6051efd7dcc0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"2d5dd8de84b67667369a5a756e45b2bbc1052528c09a97623e63dd28f2710d17","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"91ae3ad007e7c0b93955a743ad30eee363ac641007d2a656d3e862ee160f61b4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"8e2d06e2eb14ecb65557991ace2c112ac150b01caa774da49f5203b95bf05ddb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"e02860d0e7aa2efdea24015bc8b7046a8070b997e4a9c9dd8b79da285b419723","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"699f5314de1628a83927e56f3ba169bbc47a5f99a4ce9294bf80ec900470ab22","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"1ab768d3e1592f2abebb6795902a00f1fcc3f9ad3692284e2b39c94534df6ddc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"046b8e92221b0277b6d591f15a4c5f4174af67dd15019ca2016112b0f1cfeb70","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/environment.json","sha256":"6404911a2c3638834a497d6818a3a7539e6709c04fbcefd7797ff6993ebd39c1","bytes":5310}},"requestSha256":"94f3a1a4b6a4682d690d5aaee826a62b01d062403ed3cf2707b01a8306a522c2","metricsSha256":"96753a6a863e9a5aa8cb358970060622d6d5af229401d9d701b14fdf7999e022","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2478559","recipe_identity":"644f24cce52cf9030794a8fed5b953ef7db5f4553d037db2b688fecd23484a62","loss":3.732928391411746,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"198bb4d6eda37852e12fc707cec6668fc256d2bd54ab0107e80249dfad9b34a7","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"20eecdd68ebba14004ae55c45e1a895383cdecf3d15b7d5eff3748953d660dc8","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"3927f72e97434cc385678f031c65e960d3ff7ae12cd8fdcdba716ed69261c97f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"5189350d6063f2b7525bc468e8fb5c7902108378e024de726b8f05d01095d755","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"1036ed72476466975cf094c0a6f8b4734ef3ce10310fbb9c45b1f1308ed9325f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"4ebf153a367175885ed36fdefa8d7ba6b732641f8906cb006d33b53cb804054d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"83ea840b49b73eb26eba1456e8c68100e449d77868f1e46a205ac24f1fdfa2e2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"aac862e5f235fea17519bde9ad25c5eecfa657d5971bc0a214d165a5e4603e99","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"898affad7977fdb3cc1c1b68f484f7e2a4c36e87afadb4da50b19b5c78d26bd8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"218c7570198156b85394d74ba1ffe6cec21f14b798e9464626347a244dc10b17","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"4adbc74a0e29124f3cde0d072d4ce8c1f315d6b551031ce4ce4d8c2d02971b33","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"384571975bf8f37f09a1a3f562d4bcdbdd35c16e26337d83f54fa667a3e98f62","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"9179185d8835d177862fe8245edede3e9a7de150b6666e35f9e4f40592c20b31","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"6bd1b08df5dda05bcb43ea7aea2fd9bd4b2025831c6d4ff003840704bf50403b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"ff6d7586aed5d99fd250b41c8bf45c0f6b8a3f8b93b8cd4f280c0f21940d7231","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"7a2993088e182539fbfc2d2af5316bcf6393ec99b5925ccb47edad4529870a07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"df1a42d2f780abcb90b88ffef452bb05d48c50309ea6ce44d3b333b615b67ad9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"41de6361c17315983c1dcabe9683295ae3142c87a1230f8b1ef59098c663ce99","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"d029d1c1d8fca65e872d220e492b9a9265527c81ca21779e69ce9388798f8d48","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"86ba63b49c11f65ec3a618f90d1e0e2fefcfe7c4dff46f9997c9a85202771ca7","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/environment.json","sha256":"ea1d268e1fd51e302fe5908e884981ff14de5fced33ac8d09acfd401180c2fee","bytes":5339}},"requestSha256":"85a1b22b89b436a5fc1b4cd118fc088e21d11fd0c7a7f23a0fe5b4b996c00b13","metricsSha256":"3d2ab80b40933b3bd9a8a938ea94960d0410686aa1fd93ecbfb9b990a6435ad1","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2478560","recipe_identity":"164f329e060f85ef8de53bc26fc2eff9f1d2f6e968a366edec608ea4ade3a681","loss":3.759674159466903,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"5a0197d44d1f00594bcaa46865a6482fc265ac2ecc9d140561d3de3acb278bc8","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"7d7e7e056ccbb6585ff36632624c57bf0c05f885308582282fb5d153786feadc","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"3f9297c09888801a5cae15f18e21ac9bb1a1bdb9c4605d4dc30c2a3ab3411f57","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"800bda23ebca31d9ac67ea731c897aff7709d37fb8f4f38f7f97c16f84b96f95","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"3b4dfbd01de68882b9a33bf17e9e4d9899322169e5b39466342c5d158933f207","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"53c9be4e17be89544b123168f4d9693c80b9f2b052157a15833c794dd63670de","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"0924c46a195de64ec620c06572d653ac11d1e355a11cbc890ce6e066a0bd99f9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"125415face19efc9654ab813fb4269ced75babdc3e6776856dfa99e3cd50ad8a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"9fc34d546435b9c74a6a91dde6da3afe55409b4fa4a8e6933f8aa7e6eadaee71","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"4aa3e1fcb24547d3af9ba6a385a5399b140daec9e926aa2994f64bd011891715","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"1a629f99f026c2663a7d68539bbbbd781da99d2b42ac1916f0e40ac7647a27bf","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"9491fd223687c9d9c71c65325084cd8c7b8ceeed38381b767403b3ab3e9f3507","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"1c5fdeaa0cc1dfca96d42d843a485f0fb0b26d18b8e85547fef20eebbe9c06a0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"53c48df9cff2b6462fcf5bc0237bc90a080cfb09f50b31005b055e476783800a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"3f6b73ecd746928e167003cbec905ef4d6f1a7fa074695cdb8840ca946a62a95","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"07c66f559e0f5a46ec51a4f44ab42882c45462dd3e88522a69c74f2f338cc1f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"3e26f32a49a1286b05b892a3dbaedef3e1a7e62c7bbdc20c5f35aa26584f28f7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"ad7a26134b22da21199f9b9525bd1b3e5d7707642df5504dc589fdff17107435","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"baf75e43c0e6ffbefc15b0d68d1c56df81cd1d44f65b40c2f62c3cd7b20cc6cf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"426451b0c3aeb7bf70178f306016ae6de11cdb58c2ee3d49599771d30db8167b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/environment.json","sha256":"11d707413cb4bdfc4daf11946742a3f83fc28e5da69990ed4c61e6a2995c7f00","bytes":5339}},"requestSha256":"268b2b5f3b856b9e43b252f0ecdf4a9b4f588a4352fcd0b8b93b8b7359f1bc4c","metricsSha256":"6341b30d05164e00e876a3113de703aa80b8df37979b4787b403cda99bcf97c4","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2478561","recipe_identity":"e7f152d5c8309b386b1459fd261dd80984b2e568bc3f67fcab3cf959799f123e","loss":3.75243314709979,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"ad438f0ac72d83bff7ab78f2c427c89ab08349b2fe0d3ea841a8a4a5629bfc80","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"faf57f510355e550deecf331a3c7994249ca7d98427bfb4efcb2dcb5aa167531","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"e16c4892b254c84a3d860afc91a8ed5e2a384cef22ab4502ea8e9e1245633804","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"b90dd113552772dff5085714751a2ad4c6a1ed27874b41cb5416e32f82b86060","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"c339c07ee605a5df115ffb56dafda7b559d210faebe9907f75c4e31231d526e6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"191c042b1dfbf253d10e008ac2c0e94bcf348fb04177fb25d8bf9cc244738c12","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"19f9c15eaf8ff9cd82eb3d814ca14e8b49cb29459740807c4b504c4b9a3e0a46","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"0cd61624b70faf80396fa6b0e8c56ede055644cc2dd8b25b6f9d527e6515c671","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"d860eff35f5c52fee6045d566447d6d84d7b4d69f3b727638f5f3a514489d679","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"34f3333d061c44d1b79cac8b9c9dc270fa6e7b3ce5296bb67bb5e17934a842a5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"075ac3f2237437f21ef0942eb34e95f4d2f06b690fc2422bcc2ec52d146f2c75","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"874a0e756cc6afbdf46ff512c1a24552b423b0b7f536bd6efa53087f1f3e2f0f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"cc61d0126957c668e2c393389c5f8af40887c02ff434adddb07297b6d8044379","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"9cbc68e0e49c73f2d470606ebc464e39297d6583d6901ab72eb2649be677a826","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"44133824b729aa5b08243d26ebbcf03110077e7a75db0c01eb1e5ef9b2e1d68d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"0208e616d991566e26dbb8b7da16314370aed855046abd412abbb60b1c841563","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"aacf65d52cd8c251553d3049977baf2f8e8c63eab5e43f3b5b0eed018f6e0d29","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"2264359d0ee4695fa85a9b80fb0cbffddb6c28b087250d0ee4077281f60b1c5c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"f215299ff5f1c8a6d7d10447cead981b121cec2ad426b3e14946ff9b948cc437","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"437ad99a9a2ecb96a30c5c6a3300e30513a710a52f9115570c5b007a966ca341","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/environment.json","sha256":"b1ab41e729240a490530ec728ff941d98aeec6cfcc14efe6a78c127c8189c243","bytes":5312}},"requestSha256":"1b1ff07869a743700996bd807e7abff9c51ecc13ea4b2e71a6e3212fc7ede411","metricsSha256":"1b7b2751bd10e02428167a2fc9f34e5c6026d1f9ddcb9675244ab6daa11fc16b","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2478563","recipe_identity":"96b6b8c88d123eda8edde5f1d04feafc35d3c77bd9a333235a0ed45e0f36baa7","loss":3.7489837221859847,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"1e086d4fe0b98ba792da4f5b6eba9fcf1497ca4471c9689008eba90214398019","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"50261635a772488b75f43399f7a577a992500f6c2965609942be497f6dcc0c50","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"a7565d9cd5e4b41306c25fc212004e10d671aabee3d8aea9e18a9ce1ae6f4b41","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"fa82237852fe3764f8c785d233ea16ddf5bbc0ef5801082c2a48d95717d34469","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"87a06e52b6ff30776974224aedb239ff1c00add70049b680e50efd21dca5afca","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"51dbac0be01e92575c835af0a8d605e078d02de428b4689581b812cdcd2bdf72","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"3ba894262efdd7f2da709204946cc5e82a1746e4f6356ef487a8709278c99bbe","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"260c1d70ecbcf86a297300fa761d6d7c558788dd485fbe88dfd2ef605518cf98","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"1059579099db4777108a63feea93f533bc1e02299b3fbfafa1e2e64c8974cbfd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"4764086114cbe302cc570f6f2f2918315b5d6e0ae0f2b16de56863e5a5319fac","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"bff90db7c4f193a8740e1bedc764be85b355f72b6971dfc354b366eb3614488b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"a306d79ac7ca119ef5fadd240f0ae894d84e331cd30842f2a30ae6efda4b61e6","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"5759bac4a244e725b4ebd50e1b9f3fc04fc082e8ea1f5e60fac13f86400d204f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"cb1939685f8963d67c096be7ba7690fcdac4b5fe27f93aabaf377a703ae3ee53","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"d02bd20097382c1f4c3d774865e18fcf6674ad4aa4265deaf352cda87527e2bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"a50d48bfb2d727bec97b5eb0e7ee1a1530c0fc23941743c967aee4895c334bc7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc46962b5e74b71de4be42204d37fe320e9c1777f1ff5047dffb97ae37832cf7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"aaf42803ec21c0e7939c7b31453a95ef6fcc3bef862604bceadac45f7c2a39fc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"98ce069693d692b663d300d727873dacba66877fb62f788a947df1b28a3805a1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"083cc68cbad58fd5d59b88e55f9bc3b39820081bf3ca2c9d21e888efa6c769b2","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/environment.json","sha256":"b7c64ce0c7321d3f0e2646432f09ad104c926d2c46fc721bc5f083d2d385334b","bytes":5341}},"requestSha256":"ef8a9657ec19b0861119a2b66593791651430bda73a04bd60591522915ea5c05","metricsSha256":"016236aaf80216317cd16ebb671cfcd6b9871263135d3fde73c4610c95bf0a7f","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2478564","recipe_identity":"505d4ad6fea35b0fdb5d27c120a60d0c39f0ef02613f24408565de13430633ba","loss":3.780051696458729,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"179952809ca9099370c298824a9870e981c0b9882f8c25076d23b4c9b8c106ce","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"069ac9271a99b8f9de3e4252d3a4ba7a2e2b2f7db896a7a3b73ef064ca839f03","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"0fb71123291ae7c015fd38a0b1ebab86e0720183cc2f6e1395cf9546c63d8fcb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"0a1266ed7b7b3d904a8e3d3698176cb28c1b0ca768c6b1c04eff9f433df5215c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"a507cf130195000ae586030e83ffa15e1dac72dce8f8596bddf3e7e46395133f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"46589f1b77a6fe72b6b7b87f3c5353c1bddd65b26e8445ead6b5471da798499b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"304e2ebe0bdc14f83edcdd90cfdf58276a311bb7139ad7e67d90589d666581b8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"a5eb6225319a5e6784b647a0662bc07473e2878e9900ec86cecbe3dcce1e31f9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"c9899aa623d0d1ca40473dfb2e427d73a62d83b84464c6116757d655050c582f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"83a39ef88369d4ae0e38881cbfdf19860edeb6bcb0eb13cb56a78bf988709166","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"4f7ecd7a4bd8cb6b3187d1f0e61ab481bdbc8cfa4308484b55f2e7eac4932234","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"49a6dfd1bbe63f630fe9facf827ae96e7f100f735170ee604ec26c2543bf05c4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"db5a58760e1bd97d75c214c480d8aa1285834e17993e394c916ceb22968a37a0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"a0844ed215a2766e41836abd11823353dce9214a8b99e3fb20388a1a67a3792d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"f9bd7c08afdbc52b23eec2bea0f42f513ee988870181ba51bd7de99ed51f00ee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"b85791ee88f8461046e33d34a89178b70f5faf65dd21b0f05f122b2001a9657c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"442856376e99da3bf084e94f9a80e297f9c051e01d5c6faba9fb1c52407e46cb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"1e0a88fb4a1a8eb5d17143ce814e079cdd7da1d4c7c59436b8c61046f7743c8c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"2de4748d73d908417a2f1291c0a27bfb34924d18a6599e2c61fb96ccacd50b03","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"4481bb8426398a9f4550f440caea26d9eb229c5a677caccf063df93d13fdc44f","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment.json","sha256":"4c54a5e1cc632a0c88259fdf9ec85c44143e45cdb9ef5d4b43a40a056a5d189f","bytes":5341}},"requestSha256":"cf21b4e97a2d7124c921b996199d8d6005a016d0aae18a44fad636604e2e8d8d","metricsSha256":"dba14b4d4fb8d24488b166d56debacdcb8b39b40a966b3fc1eaace4988a385dc","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2478565","recipe_identity":"126e62c588aced9e33b09ed779a231d459e4f6da554815c361db44a8ee3491a8","loss":3.754887184131928,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"8198d873d9836a3b4cfe7551020236dec5eb34da2f0d033830165ff054692698","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"9716ac7b481132d10650fb7cefa34892db3126172bfad6618132fe37be5a1079","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"a47173692142f21c5102841b19c4fee54fbb8ad8120a88c640e53249dd176f39","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"28ceae328a9cab0248e687539e11ca9ebde12e33454e78f9a5da8463df353b46","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"1d083b848344b7723033cefaee959cd35d89ae27b195552475565b1483d395b0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"055a3c8117766e05909d43fad6571194060f7531686f2a63a8c059d6bb8bdefd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"8389d1f7c2ab121e00f713f6ceabe0d9960c01eb05fa6925f27c34b2449d6b89","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"861f69efbf1f5719d1f0223e0b663129da3561ad949cdd7b9d8492c0a73cd6e8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"770b53973fbe15711e7706089cb4cf31d97b968a85fbf9094a40ca69f1ba7c99","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"d976792a2358fcb88a7a0604e1ef082590480c05db5b31ccb507f996b421238b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"2a53212e43bb82093683e621442724cd1b665e985594694f04494b91060037a9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"953f4411376d3aaf8f3a7b771f542d54f0ed7e35e4a8bf4372b212337ee5a391","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"3ad82e13e39d13be2a6b490f047e9bd8d4c9e333abc7b7b71d83033ea80c6592","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"2f1d398bd90983e8bbe004236a1f80e29b801132c5222f18fffa318d6d2020ab","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"a1258a961b01fdc084517b034e4d9410f5d612bc9a20bee0e389048faacd8096","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"34080b8a7fb971b5e6388c80bfe89b8762d01bfebd8af1dffe40f91fcede3188","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"0ed0b63bd7ac289f56c5d9673cdb5bea98f1845e185ef43967fb475b5027cad9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"a5a2ab931637383b3bcc1d6bebc34c495066d0c719dce24189cd3c97afdd1526","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"37066bedac40480c32f0e1845cf454488c177c69fac160c9b439a48966ac6cca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"ebb6a591fcfef068b8fc9cf6347fa9205a835c80374e9db057fc4c793b07d98b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment.json","sha256":"502919715d8fc887b8634e333fb97354c07488357ad3ed6d24414768848f8b21","bytes":5312}},"requestSha256":"b96c24d7d69463dfe593dbfe8ab042223ed90b96e567fffa03bb48e2339c5ac8","metricsSha256":"4a58aa5a56efdbb5fcd1894a019cd0d171786911584d07bbb89166d39b49b8fd","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2478566","recipe_identity":"740e18fdf93eb26f3f127270de2b77e776ddb481af2ad2a935a9c05c720e8c05","loss":3.771339394482213,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"8f41ad843afdc43a6f70f10ba159afa61e31fe30b9a78edfd322e99601397799","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"32e45a47f246a65342f747ca7ad32e8df16297a74f8a73e1f00da1df6b3ed624","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"ea05f41c67ac54589d53ef2300cae633376c45429eecc9850ab1880cad0c5a9e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"7bc44fcc09cba93e6aef88c5099c25c14baa1db6ed6575ae9e671e4e5358a2fb","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"6beed4cf3eaa7ccaac85056930fa63bb1f918c513bfc47a5f15eb413cdea548f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"967deab94269900e59578c5d12a4759bbd15b260506ecda63a85199d50604829","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"5dae7655c34d2f455feb65e2b9a6e96c325911def8834711e51356df28c1023f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"c93b923baf1250ca6ffcdd863781ae2321cab9d3a73e334ee97b7f0025d80622","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"1c51083b550b75181ee1d92f4c999629e40bcb06348d48a2a28fe1b384ed99f3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"43f0a661fff952aacd941255dafc6f846ca5e4cd5ab909fd794fe159f3a46e5e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"4d30f09e6fa2a1d21a6bcefe78865559f46b275c152583fff382e6c23d690266","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"2851f62a7cd4ff003b233c4cd9fc20a0bdc96ba036fda072ff1583e2fc1b9423","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"84fca79fc1b0ead4d38f68abbdd4f900337e349e8cf3b374eeee46693348e289","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"0d9deb29232a6eeb9f5cdba4392e0eaf1d243c6d1da29f2103f1b42c286a69be","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"0f652019bebd2fe4c3b728e04e127f0fc97a62b4d1cc8f0507f3fc50b1771549","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"813f5692f716b3c1ac9485fe9ca51abca8f6aee33a1ecc8df9a960a08682e67c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"e5bc0ebda6c766401cb8c9354471b68a04a0a3ee308381b0a0166e4d26ca4cd4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"4d8800c03c35217614ed5055d1bbde38a6d57d4c1cd2f4741962c5d7321c43ed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"434bd8620f9cdc47a9202902b404dce023602a4be37e8d23bc9898545803d245","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"2645bdfb03f6bf226566b8f6ddbd99247f394ad2aa20b545a130c4dcc08c53a1","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment.json","sha256":"4f5cd1b31ffdd4423ce3ea9eb8b63bcc450a93316273189a9608a02fc8322ef5","bytes":5341}},"requestSha256":"658718f7251a011cd571ffbe56ceeac477416933890d45d51ff86d22dcf6c4ec","metricsSha256":"3b059c8dc89c7f7390de6381633f5bbb22957533fdba181500316756fd73b6d8","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2478567","recipe_identity":"fbdb252a29c004aea59977f2f7c71c3fac7de8155afa51e4f3a30221f1354e3e","loss":3.7996437607319002,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"1e4fea7dfaa779b587ace682f1637d5ff316688cd712ab4d77e9b12a4ecbda69","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"f91b9af49744cb5537798e781b40f9acc0574429da16c076991eb3a42f1119b0","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"d43ea03e01bb4f0e6d177530c38d0fb35c765daeed1e97589c46bf78a490cb4a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"7c80b1a108858c7e40fb36f95bfa13737492f7a5e97a5f678beda2ce5f029adb","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"2120685525ed5da3209b304cd2cc9db0bc12350fe334bfcce728ef2df2e733cd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"c1fc4641cdaa7475be898e06de6f074cf6956e4b3e9f9693097c7e34e5da7c63","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"571d04c2d578c73a2666b25cf0c596380dd38b427d03408ebeb44c09e0ab7c34","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"74254f9aaf8972cd130cac514b7102b99788f435d6f0d57aa2154da932232a3c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"bc7d185cf9b92b3ae605789b5022d20ce58661827d5e0208a34ae12a2984fa1c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"8b28cddc58e48d2897fc92d451f57311d42eae38d7e8f0aeebaa7cabd13295dc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"d75f98c74add5b8a67bf03b1c4463a756cb4fc4985c45432ab117b4e164ecca4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"c3afbd40084ab48a65d9223453ed2d702513509aab4787358849ffd26f4dc11c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"7476859d8b0717888ec2bad9c2523fd314060b07a08c512ae28e3ac492061bc9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"5d364bec1e497ae32148b72ee82bf2cb4cb35fb852462efe6c81b2a437d4893b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"ecd9648684b9222c4a5ff92fe3ae574ac630bb939af7a7fe7b493085dc6a1357","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"7dc39ff7f5122ce58c9d83476ae72f3ccaba54f47441904623db881c8e312b2b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"67f0a88dab993b161ef661c03a536adb1acbf99233bb69334318ca428cbb6cc5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"a9938d6373c5ae7e63b01f21f5261d3221dd9660a6ab0d794a7d657d9d1be278","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"10f9e036a0427986cd1bcfe769ea4c95e94215a9ea05e070329d003555f91c07","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"b7f42d5e700dcb6a71b0968be16d4b85d4f13f29e8b151cc4d20dda735807247","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment.json","sha256":"fa03e84a018d5e43498a6f0fe7efa873c499a9f2fa50cd0b6011ae9f06943365","bytes":5255}},"requestSha256":"4014d8ba16e29496f66e77c05330f5748ecb00da58d4b6c616d0673f73d17743","metricsSha256":"81783ac5f59d6c8c7acb9619b13edcb3ac768fcab934d9226ddaa4d115d4f411","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2478568","recipe_identity":"1c0ddfed5a4208f3b522085649b3af43774c2d5da9919d67e30a699bc064edff","loss":3.782585271397604,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"b64991bcd23c4caf3da5f45f8d2ff43d54de4a4c5ae8963975192f8d5ddfc048","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"aba8a04e15bd3126cf54d3e007ef5370ce391eefd551f0e62bf10a06feb4a0ca","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"6af02c8ec197dfdc4d79f53db9490544c94c7f93f38a23eda29e816c89f6b4a8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"a833acf24acd2736c8ddbed7200cdc8d1f9890c7f7c4c3574faa0a69bec8c2bf","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"42e07ffbe00f2599056f7c8c0bf83b79b81a2e63e5b07c1992e78bd44474e05a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"58c262ec8b0df80017305a9f16b46c13481ea9a1985d7d21695c8690de9869f2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"935aba63207e3e441895399016dc9a8b133885c51ea716a202a077f896b92bcd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"481e2e1f1134631d81f8a21ee8c189f36c5bb512570c72f448ae3eaf65990ae6","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"c9bd589099dcd0c6a90f22ead30f27258522cf94541fc82df6e18ac93b375c3f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"190fc4733574902cb5fb9e0dca510982e6c2cc9af22e3cff229a826171ed70a5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"f571983ea69fff6d004130459dd4a2aa1e0b64413eef3ca895d2f55f2166a041","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"81811d4d22627a10cf1661706d5bc180fb2810835b16185e9fab075716e16cee","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"28b97e499d140c51c960251a3a4f7fe898f91ff4f75b84be97baa7b4fac8cc70","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"f3abc7aafbfe5f8a3e32e069827e762fc37d2b598756c0b0bb5bc7dd5bd711bb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"95781084f820dc698d1fb632422873f3280a2b77b39b37dc42499b1d63f107fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"abce995de5db1b6d39996b7f0ac35f0604adb3c8c434394de17e749204f853c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"e36a2fc4934e705e687b4538ff1b0b2b726b681930257feb28cacc953da668f9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"ff272e2acb91037fcff7a54651973514346dbc150dfd9bcc6ea9ad1efbe9b8d5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"f24091489393f4f6d8652c1ce6c3b63119999de55621731d0ed6e70e057a0d9b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"0e1ece2346bae91afffe59f2861f9b99e8386425c2898f149dd45bf10de22a2d","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment.json","sha256":"620e49f2172f491f0f5a169a603ed1376041f1e09b931d7adcd072e64b904b80","bytes":5341}},"requestSha256":"758d6e7a44360ad46d0d9ce16cc7b710706da6989c96aef614a4a43f7e135479","metricsSha256":"cdb073f0f91cb6e3940a9f983c225eb8fc4b8ae2d3c4e6a1d01fd7a0d9dfc676","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2478569","recipe_identity":"057cc957cdee874c0065a2729ebd9cdecf4c4470f182181820a6907dfa54819f","loss":3.79766998233501,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"6e78542e32e020bf953b09e58fe2f39b21b447de433ac97d7a90d9c84e8e820c","bytes":881},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"1c35eceb9dc2d472bce8503b62408c1d95bce3096db02708c12ab886143cf271","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"9a01d3416006322408e47617f6431b78f2e88d9c887bbe9b16530ceeaeae4185","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"b5d91f71cf5aa3b55753b8e7baa61190e331fa2f8ca6f9a33a89c1e460059dd1","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"b9625a15b5ad09b5671b38def0033578302953b88802a8eadc36d59cb9d06f23","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"6c92580228a7cec2d0f93441aea41e8f8aa4a79b0ef868e8faf0bfc7ebcd8712","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"5af7bdf72c93cb076805ec303fbb9f708d629a70a325e1fe3533f96d4566614b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"fd89f292349a251c9153d43a819c0d2b5e45f289a82920f7949f8999bd6dbdd3","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"03a78c87f81a3200a942300fcecf4540cbde56d844cf7d567e5d2e0e84042baa","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"cdc2d500b81c42ce505c53d79fad18b46a910ed063d48921a37d0b219edaa797","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"3487d18b853cd0a8c3ecced453193ca4c83bb46a27a78bf071423092e3909f88","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"8c07a4c77c29afabbd57f914cb0bca11079071fd297d281b11d9081da5b5e481","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"9d3fb32fab63744000a144827a21e3542912beeb520f1cab54b7610c206d5fd2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"f47174f46fb6858cdf4e8c23f5175aa01a2b1e075c907b15e254b2e03f26b238","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"2bc7540ca9064135fca86a9d750ed86f4fefd80d8d336fb44f6f8ce8ac78622b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"912b4ec512b3d479a1ab89c22dab6b9aa4e7e1fb1ffadf0ec7d4437de1a45f14","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"c612b8bc7c2ecab85a0753c339d98de3a817b7a90efaee28ffd5b96c48fe0f22","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"71bc468db407711eb2d77c093f0f7bd4bc6072a0629f0114e91c0f11c9b3e630","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"37fcda7c788bf1b8b24b4457e4585cd84ba7b8f8880267f08b7d998145cf91e3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"4389b89fa3305d07a21fb88f4500cc67bb3ed1cb00cbfe07ab2c835e0fe40282","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment.json","sha256":"3c70779a93c253054c860fcffaea7f0ab7a161b80f305e6c36495bf8b2611e63","bytes":5341}},"requestSha256":"e9206f5c3fa47e3aaedc5015469118572d9c343af75d37125f4d9165bf18bc74","metricsSha256":"544c1bfd7faa6e134060bbf30ea5886c585a5f482162ff42b8f954f66c9ab2df","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.01-s42","horizon":20,"attempt":1,"job_id":"2478570","recipe_identity":"ca0dd55ecca8f65e346b7b1d3d21885494d200e424eef30b169b8920178b79aa","loss":3.649899225139402,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/result.json","sha256":"6e1d8c71a29aba0057d9609ffc7afd33c21f72af4d0d3bc70473433317233520","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/final.pt","sha256":"2c696067b427fad400fc2716002da6fede28dfcd3eb0c05c5878164eba9ea3d6","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/final.safetensors","sha256":"bb48e2c251024ba5d24e783730bd77c4615101a5eb04d8e30f68a447f78b1c0a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-000/final.safetensors","sha256":"afad447cddd265ad8714386852b73cc8302584480de4fdacc393dcb68cf0aca5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-001/final.safetensors","sha256":"9e9dbcf4812560baea870622b87e8163b63a3db5361f73cb83b6428358fd9937","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-002/final.safetensors","sha256":"4b80708b9f8b96efc6c82c88192c818e8c23d3ce00fb86c5182c0aa5cd07da70","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-003/final.safetensors","sha256":"f04b8f1c869da71764aafcaf483f33d2dff6cad058dd2c31fcc9c6d8a4d9956b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-004/final.safetensors","sha256":"ab88832f8cacbe1d0d9b1982524c19cfac910e187d352ec04f1094dadd9aac0f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-005/final.safetensors","sha256":"e902839041e6fe979f706a06a200924232c3298555381c722864320d59c63d88","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-006/final.safetensors","sha256":"b8e200f6fbf8e472b72c82e2426e30acf5aaa0dd0ca9296a6187cb1af27ebbf4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-007/final.safetensors","sha256":"123f078e4ddb0b6b899d3ac6fd011774c3c691adc6e97a9534db99c57d54d6ec","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/epoch-035.pt","sha256":"07a14b9be6fc9c64c1f7f5620ecce1e61d61c58b18d8ad056e346c055207afae","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","sha256":"b0f5ebfc13563936b170e116ac2b07ce71afe68966a37b55acfea49a7b98632f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","sha256":"541bb427968895f76f90af26257bc32e2cb1764570caf1f40bbe986da85ac268","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","sha256":"b5304c8f5d988416f54a36adfcf399ed5bd20c0592c94f8cd84c137e532e3986","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","sha256":"219d6aff3fbb3467332992d375e955c8def96bb5636cac3330959922f192da13","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","sha256":"c7b6b1224d1c754c43b79a341c34f372a21386cb8bec027827d46c3c9ee86acf","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","sha256":"3ef012ef303923f1cd93b6f3fda847a4a5c7029118925876861eaf445ead2f29","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","sha256":"f380948468ff3ee8c92cc1c9c1e41a33e05a48d2757fb99a74b2dd4694b1778e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","sha256":"55a76b2c24bf7f5ddfe69f2132238a021baef92d6c06832220b42a570a21f359","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s42/attempt-001/environment.json","sha256":"6b71d275998f1539db01c3668cdbbac1c610650f5f2b338f06f7464bbd3b6057","bytes":5255}},"requestSha256":"6eb6e6fe20ac8b2a6bf99519b9245a708cd52b0d959b66b3458e12616cacaa05","metricsSha256":"dfb6da303107da5509eb6a061611b91768be47aec859b606a521816c8218cd95","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.01-s43","horizon":20,"attempt":1,"job_id":"2478571","recipe_identity":"469326e6ab25253f555df041b4eac5965bdac2b08e763ff1e8c3a5a927e6eb4f","loss":3.6421674317637716,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/result.json","sha256":"9cd9f887143c5503bb6032197c1cde07af9b7a3706405d60077adbc44b822320","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/final.pt","sha256":"c8bc9154af3196b02a3ff39488289990ad7897527b8c19f03a366cecf5850112","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/final.safetensors","sha256":"766cea93b8eb8e3e11be0246564fd3aa9301a257f8328fba4d0ce8fd87a66fdc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-000/final.safetensors","sha256":"3c56c630bf5a009003fe4a75ecad8d7e05615bc1021bb47bd5eaa20dc527ceec","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-001/final.safetensors","sha256":"ef9c906ec500926b21307db7d953287524188e4dfec6ac7207ef7930a2483886","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-002/final.safetensors","sha256":"482a4fa0149d64f1da16e375412d43b641fb8aee1a88485a88964fd208a5e356","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-003/final.safetensors","sha256":"f23ea15a22ef884c8e8f1aaaa10420db0c73a231462314b158a18c1334a58609","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-004/final.safetensors","sha256":"8a400eb4510bd32dab94d370544c037680b6b3914cda52f078ad4f6cf772b707","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-005/final.safetensors","sha256":"98d46e6fa98d80df28134221e39b8e7496798981db9a2a4a7356bfba7dee030d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-006/final.safetensors","sha256":"d89874bce4e0446581064ee8f4f852693887c02dbd40de00de41cd90411f01f0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-007/final.safetensors","sha256":"ad44aaa9281d7e6908f0c7aa17abff1874476b940a5859dac8d2119c305d7536","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/epoch-035.pt","sha256":"9562071106415f5588fc9511d8ef744d97d21268aaa278679ee9cede5e6d7138","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","sha256":"1119faf8da3385f5e5182c701fdc16983eab29d31a7ea364897cc7c2fad5e46b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","sha256":"0204adc522fc1a019234650737608e5cfea8a3f3b01a0a351d0c4bce27ac037d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","sha256":"73a1a30b57b9aea75e9eb6240f67316575401fa76ffb768babade966517c3aba","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","sha256":"b48013aeeb19a05c68d48777a51ffa68f6146014865b978f2ffc055cf218132f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","sha256":"8cfb71c02f4c76b0022db69d5a4e774da8e19472280a61e9f2ab783b361c2b92","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","sha256":"0850e707e21d357aa4d69b08d7511d8faf1217c9cfcde53dfd86b6315349d323","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","sha256":"db9e0e7e75f038a1ec7b33612d9ca5c0cc951c61a745df924aa2f427ea3c933b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","sha256":"19bd41ba2e698afbeddcc9e42651f3ac1edde567e46a1b8adbaf95379dc6faa9","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s43/attempt-001/environment.json","sha256":"1928fbabf9632be7ac91b863c2a08a4fe075c8c01f9401cfba42528289b5657d","bytes":5254}},"requestSha256":"6f037f83856d4fe337ca97404f3d2252e4c5aac8b46c67dacfdb61fd9b82437a","metricsSha256":"2fdf46452c168ca2f5eb4793a4448d02dc7319258b6421cd0bd741f4cc225d13","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.01-s44","horizon":20,"attempt":1,"job_id":"2478572","recipe_identity":"c38d133e3ccf5eaf165ffa092691b8865ae9718655dc9bbf213e28aab7d0cfbb","loss":3.64566813186026,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/result.json","sha256":"26d76f1c258f438212d739cf5724b8a9976084137cc3e7f6e0ae94fa7356a507","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/final.pt","sha256":"cb82a423a59cecbb93e2ee6bb7ad7ce1ec75ff192429cf0c6e9fc0543badbe67","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/final.safetensors","sha256":"efe14e31b0b75cc9d6f62c2e32ee4884b8e893f5f72c4a94a5a8eee08163b2c0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-000/final.safetensors","sha256":"7e530e40f141522b9bcbe8d40b613c3f5b21477935f95bac97793bb9ef298eb3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-001/final.safetensors","sha256":"e9b3929290c255edb6ac16164ea6b0e2697cbd389756df8a208aa8195282e18d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-002/final.safetensors","sha256":"6d0b8d37fabd74a492e70078d55af138a1e38d700c59816d86925f5cc395a79b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-003/final.safetensors","sha256":"2bde505b30923daf02b308e0b729b2b83eb87e66d8f33863aaf256b0806fa5b2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-004/final.safetensors","sha256":"67f093b43a44122f521f6ce43456404d5f5708813351d6f340cdac5d976748c1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-005/final.safetensors","sha256":"7d0d302114664456905737e5662ed99f3b26b6cc77b4e3e529567f1a322da040","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-006/final.safetensors","sha256":"6d53881ce57e601a53b121443543040b3b53d9a766c94f42c9b0249c55e0d47d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-007/final.safetensors","sha256":"176eb24f3ddef50b122e7bcde0dc60ecdff0ade060aecf85049d962621cddfac","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/epoch-035.pt","sha256":"9c8065eb07b89077936936b3ef01132eac96f1d4652e75110c879ba489933bea","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","sha256":"dba9bff89525988ebff01c6f3dd7a9d5a4a060ad91a46e1f6cada264adf7d831","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","sha256":"7cf1b7ff3fdcfbd64397bb5cd15429daa1e7f5ea292efac9170eeb63686ad24f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","sha256":"67c4c19209c724773a5ef00c74d24354743dfbb8ba714c9ade2f1c340e7f66fb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","sha256":"0614489e5b35f1cb4951f6507b7c154fd60933453518053d6b16edd70e2f72f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","sha256":"2745d9049b39315693c2fd5c32404c4bd90e6a251da2d02c1db776f866854129","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","sha256":"4447d90667dd5c9f4dbe22f451ad986a6d80878e5839cf97f1066f1e7e7f3d9d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","sha256":"05be689ee70864b331bb44269efbe4d029ede80ea96757732f21f1ad68fe18ed","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","sha256":"dcb3b5a5279730c8e970875022defc29c74fc8147232e724740b08f80b779191","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.01-s44/attempt-001/environment.json","sha256":"89e5016a2979d2cf29561c62bff4d4de39a0847e90169384974cdabd064bc9ad","bytes":5341}},"requestSha256":"57bcf92be580e6b358e23240911c5a1ded710b89868727219a9e775218f388d8","metricsSha256":"0edd4174b5c31a9308446caa2cd01c6f861a9299c265e81ad7a9b9a0cdc767b3","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2478573","recipe_identity":"f0aa37f8934b23a76b65e3637e65c81433f389bfccde7b05c9aa393c341a3179","loss":3.641203605182922,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/result.json","sha256":"90336ab088555ac9ad7be43e5c601dd1513259f28d390bff8b7c1a2872ac20e9","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/final.pt","sha256":"3e0d70a72e309994e0bf530b46fba2d2e418b5fb5028904434831ff95460298e","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/final.safetensors","sha256":"697110cc1dbdf22df98c52154758258ae4f31875a1232f677acfb101dd3edae8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-000/final.safetensors","sha256":"7ebf3f178b49672025272c0d090d689171c2e9cf8790f88fee4c696a4d064607","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-001/final.safetensors","sha256":"dbc166cb9f0e95e5055e1dd964d63cb30f8996e8534f1f9d427a2cd405035ed5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-002/final.safetensors","sha256":"cd01941716c1e4a1384862c9ca9701757c07510a26d9460a44e7bba12b091fa3","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-003/final.safetensors","sha256":"b632da044c69daba3e0e66c730ba9087bb024a0b3c095ac6db7f2d7ffc0f1a72","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-004/final.safetensors","sha256":"7acbbaf0d475fcbb2cd334db4dc876e2aa0f841601cfd7b0cc329e0e21375848","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-005/final.safetensors","sha256":"1f8d99fdbf681997e5e03083d40245c3270c4962ea07dfd831b13dfd50353695","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-006/final.safetensors","sha256":"8995dacf2f44e1a3b0139bba63cb015c9ff86c753f2aeb56bf08c220bd644f90","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-007/final.safetensors","sha256":"572048b8237742016a37c0cb4e2af81623069ea9040b15b1416eaed9043929c4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"c1f4ef5b3949c6f20ff45e5ce81ad467d76c31d710cd9815bd703f30330c167f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","sha256":"f761fa216653446de89780b10b9a485679a261d1f5bea8311b55262422b7dd7b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","sha256":"dfaf8ef2abbb6e2f38b66a4f44fd907a38388c7825a2354439c7866a5bccf571","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","sha256":"1e4cf536e3505b1cff1a5b9f5a6624d8ede7a48f0e7c5a80edc05cfc2624ad53","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","sha256":"b11d5260ba71d523f4f8b01886948f96750af55886222cd16169f9422b66ea47","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","sha256":"633c3d1196cb1526280f34b14dfb2d9173a79569bec5d397c10fbf4aa3917acd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","sha256":"40935aa09f64e11881b4df0c7644a1a1be2e510084cf6db417d663e4b8f32e88","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","sha256":"2c804667b3f86b91cc0803562e0f9d4d344c70512483830d2a8a50ba08d1c92e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","sha256":"da0f66ea742041e739eb198691344921655ff9bba8956adbc8e76edb0682eb1b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s42/attempt-001/environment.json","sha256":"4ba5191aed25b0d8d5b4dcad28fd08b0d60ede7939d419fc6b0eb38e481f626d","bytes":5312}},"requestSha256":"37b54d087962a9fbff850fb98a3a5a987426ee5e9c315e20b33cd5a51a224f42","metricsSha256":"37be71b9c435b2871608de9d36477e45a839ce4a629f7c97f8c7708e11637f0e","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2478574","recipe_identity":"89b14fc31e96e03c7205d06464e8a6364620752df08efe84b67e5c1da9858895","loss":3.6390867063940076,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/result.json","sha256":"78fc108ff54c1002766a5a0e15eef8c91072c5cc1abfe819ad397277b7d534c2","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/final.pt","sha256":"fc23d431d6d2096b29361fec2479b47d337c6d9af3c9004c3d1a4e3936042984","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/final.safetensors","sha256":"f3e36a9a281f95273b311804e333a140277cf5172b00a6ce76dd9a41ec44b51d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-000/final.safetensors","sha256":"f0175d06420533a35aedd6ed558f02539d898c11d4e0a51249f03910ffdf4591","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-001/final.safetensors","sha256":"94a3b770a44700fd6d22ab8cdcdd0549e7ad3a668d4264250502b951ebe75917","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-002/final.safetensors","sha256":"e1c3f3efbdacc694d1bb823f0cd578cf5191c5bbcaf44e2ebb54b6741810776e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-003/final.safetensors","sha256":"fd01bcb571b2d101b0255339ab20502104fe4baf4a3083192875b62763ba4a10","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-004/final.safetensors","sha256":"d94e2ba91316d2f1db3a78f8a094b0da888303f1f1ca6adb4565d47e8b48f199","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-005/final.safetensors","sha256":"c33f071c7a5a3a962e0cab336ecd5904c02eec1e72ae2277acf678aa20414e85","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-006/final.safetensors","sha256":"239f13f860469bebcbce64ccb5803d0340b9931d862f12f63b975f70c9783b3a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-007/final.safetensors","sha256":"40ff217c76c3dd0679878bd8f739718f15ae1be4c99611f6bbb73c55b40e3845","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"d4522ba8c1290192927bb7fdff4f881072d2a973cfc131da31dafce64f536009","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","sha256":"66d8c05517ab5418c5a8b743a6f17f771c9d83150a0c9078402ea7cea0ec63ac","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","sha256":"173d89395541bcb175b1451da5b2a6f46dd60631e84866395527b7d700e22479","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","sha256":"06fbcaeb6668228f6c8adcf207785b310192f218e64a2ca76643b91eea9dbdd5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","sha256":"9c5703f3fa555cbba28581dce494f34c15aa000b3d57c5208124b207378edc0a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","sha256":"63cb3f1296d65ed590affae95e2a112e89bb85d1bf9d9da2ff11843dd2527438","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","sha256":"b0879c08409f851f983264d6c0218b07fff8f1073ab0c70fdf961a0ffc93c6de","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","sha256":"e2d86f0d0e15c92457c1c07f4bfa6ff652d41f99e786b5a0f56f6fa32248a4f8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","sha256":"17d0728af380e9af42a9a4fe33ae21b2b83c4fae3a5d4a7865ac3dc8fd2c1108","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s43/attempt-001/environment.json","sha256":"90a8fd6b059f090d9700c3eded78c746fe2e1a218540cce1368aba29ceda41b2","bytes":5341}},"requestSha256":"226f0b55022fa40cddc251bf7fcc6936828b7f1fe0ab59fe2d26f29033881465","metricsSha256":"c24e1e3d1947e064fc6bc0238890e20245ebb56d9253724b5aa04c5b413357f6","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2478575","recipe_identity":"9937a39fe89884c8b12eb87aae5c1f6c2a462aaf67754d0f6bef2ddf7fa62619","loss":3.6430117021673647,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/result.json","sha256":"75cdb8c16a06865673843be48f92210b4f801f7a7356983989d2f738469bd8a2","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/final.pt","sha256":"ed55547b392cd52a36892771e45731015e568ddbe6b5cfa49324821b48af47d9","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/final.safetensors","sha256":"3e3637cee3746dd6ef92cc452482add78cbe3d745ed06e7a9916ca857939fea8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-000/final.safetensors","sha256":"27620a43c99803ee08b85f5f8e2b7665932b0fffcf8bc5e82283c9557cd35b9f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-001/final.safetensors","sha256":"38fa8e5400986f3646e77cb8b432250693dc8e379a28c39d59bef1e85090ed7b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-002/final.safetensors","sha256":"415daf8516161cc891aeb5fde8f5f076771a04fe7093f77e062fb6d685443950","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-003/final.safetensors","sha256":"09d30356eca5e0677d773ea97012f8fd0cab7530a51e787ee5c5ed983eb001cd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-004/final.safetensors","sha256":"026a3c7f496c65942a31374fc3e8fc5f559fddaa2f8c679e79983d630cfb0f42","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-005/final.safetensors","sha256":"0039f7d3f818f563be68b4c803ab9f677b2b1f0168621d2ec01e1f519ae12437","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-006/final.safetensors","sha256":"8549da86145b79b4abe4b14eff68222b7d1f69d0cac3b2af30ab89a9896701b1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-007/final.safetensors","sha256":"2244b9e48e851afd00fcbdb74df09187b679534704ef44fffce3013c365c00e7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"8cd5a5c4a430a19ce832d996c69b7e753d25fbb1fe5ce005dbad03dc9ba49362","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","sha256":"38c6241d889844d3001a2214900c62f98df639c6fe6167e2458514ed6f745653","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","sha256":"ddeee9d9de77ec65511c7f6664270f69e38b8caeaa1b50d5f7b373e132f8e58f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","sha256":"dcd8d3ba5d3616433ad39d5d07977463e1174c553fc24f36a0a3d8f392358a02","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","sha256":"f0ea3e32a45c89f6c5eaed959ba69858be69e2553c78a7b2e673d6f8d69dcf64","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","sha256":"b82f93a2906de0b7f942a9356f4a0ae01730616fde498dde34c33c79b24294cb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","sha256":"cdc07e87ae039b611b2172c955f0291287fd0a778b8b97fca06c81d7d2b25639","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","sha256":"4f359b54a27302e7779c8fce6c01157ff971d22349c04d2446164202cb75c1bf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","sha256":"5d7eb16c5e8deaf6a61c655de66d2f8586de93ca1f1dbdf7177d2c97e0b768cd","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.008-s44/attempt-001/environment.json","sha256":"3a4a88c515feeea2479bd98ad57f2fa0b90a3b8aef488106c7b4a4ace77387d3","bytes":5255}},"requestSha256":"3a5f15a66e8f578d97e62cf82fd440e179040abeeb4b6543cf1b49e1e99946df","metricsSha256":"84dbd43e3a6f0c61e1b34743785b13d90f75ec4ee966f2616b2f7f5fff410144","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2478576","recipe_identity":"a29152e224b781b11bb46e27222621f1bd112b093d0392208bce0f98de87854a","loss":3.6348810709061317,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"a4791ec8f8204bc6e854c6c586b03ea888cff2581872f7943198a6d544b4372e","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"e037945c081456ba650a7b7f6eff8d7458a685de4f7d80ec5b3a834302cb07cc","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"509ddbd0901918bcebc8dc8fffd7b8c29afe82810cd412d80a6af8cf730a3c00","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"b9fa05e561be0af95713201f2209381ac363db82449bc3790df926d591c86243","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"2152b0b860418a0026b45046fc443fa334dbeefbcb72238b25843f2974350be1","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"d1d2725da54f3357f642c656d813b3eb23c8d6632d63f6daa6646e0d3915f904","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"0f37f0131de1ba30dba563674dabf8139fcea06389372e851aee93739e6cd9de","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"9539cee37f21a6e7f83395b74dc281c8e88ce64b7a01e22de515dc93b1003965","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"7b61467deebf076a3c1d840b5ef194512b9b9c159264b7d5d41872c414e177e6","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"7cd8ac01e43ff9885edee98599b8a9d5eedaf30588813f89b4f7d48be629ca64","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"650c6a27ed122451b1b40041f7231cecdcd9f9fea3477b9548d93bfe509155a7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"cfabcb9e5f0fcbe105c75c43dca9b56d984089ef3b851f4c2676e94636f5fedb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"9b37e29d973f9b0c7ee6fed3c78615882302ed4d30a73bcdbfbc48aaf67f45f3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"e706439e7feabff4a96ed4b0930c9a191c062c49bcfbf28094421c585aae50a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"9ffa0c4dd14123fd1a496ce40a5f2de4bf280d5ac61b81f321e279d76b4af240","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"9ce2e2be8afc4e257c9f748f061a145a28d58337bb8e46d0c1dabada7c8d2b72","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"102f9d649c402324d90c37f611bfdf421f959ab54fc29a991bae68769fe2c01c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"3dade48c26338f4bdd4680d566e73c17bcd6fd28a9077f75e3bd4bc882edaf17","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"21bfc3a0cbcd0a1d6c5618dc8a7418f2d7cca7e5a0252e4c335fe65761ea76ae","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"da3ca290d076e5ddf464b55a2e203f884f434ce594aea59dc8d52ec03ced5798","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/environment.json","sha256":"9f26cd6c871672ff63dae2388f1e45648fcc56fb0c03187c17013d93a661d5b4","bytes":5312}},"requestSha256":"d399fbffaae25bce23786ad335e3a6c371d79453cde49ee58766e920fc0bd22e","metricsSha256":"1c70049718190aaa15ab670365efcdb26a725b881e281397c9780976a3e33340","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2478577","recipe_identity":"f8e540bd529d78a9aa98e1a098e25093fa316ae3e501786e893c9fd29511d3a2","loss":3.64469187089275,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"f19747489456ff0c80eb8cb5c2af48a2c163524d4723b0f166f484321e242cc1","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"957c54b7b267714cb73c8ff7520458db5360d82e1ac1608ad0b5698397d02f47","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"2b604b568521ad4b176de4092901e3fe6d874aeaa26965f5b956bd70555114f2","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"6ea487bbdf6eaf49d0e784f82cb83d6a602709d86efd8fa3a8ceae5090c81ccb","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"bb3fefe47d8fc4cd36b69e85aa714f87bf7d711f2cd0e6504030cf6227142eea","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"a73a046b67dfaa33eb99cb1f3a048838e9c27cd16d14d7ac1785c3245791a01f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"c51a3e95c32e05f24ac8d3791d61bf957c590aca3aabf6ca9c6ac45ac1bdbc1d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"278267be2d52e32ba9c814c93b69be3add3f8c50a6d4033d1b7a1386bd788138","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"c4796ed271692c4ec600b13590210e91998bfc511da4c3377c5959e6f86bbcac","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"6805a09f51838115f09c7cc6a97679c7b1ff31b6092b25608e216237f83e2af3","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"fd3266e9729e130d94a6bab0ca0a4ce59948c7f16dcbe3d1e4ba1b4760a422a3","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"ca2d0464831f1691f1dca08382049a17fd5b322005b76cf13a3f76cca46223dc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"23942e6c81b1f3fc6672de43d912cac3543b43ad21314e6f2c4484d7577b6daf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"0d2f0023860ea739cf5caa70ab52cf4b086b60842f8d4cf180dbd261b2dbddb7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"367dbaa627aa1da7014db85a404cadff8e5a85ca3f9440b43c3c407a1df3194f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"3f2c02eecde0cae5e5c7b3c97698f1e4bff464f2050a657807970025b7ee28c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"e44607361df7d3f8b51a8bce643c2eb27b4d892ae991d6416db6972ad6c5c034","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"d6850c4302534ced88625b67c6001a185427a9798b88b8745e03db1bc512e3aa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"ed1beba7747087db91a6cb6585e956a90e0fe11c8d7e2a0a13e1b18db2e47b71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"2506a64f3a3abf73e3c07c80f1ef3510076ee793f6d282320bc40a5370e6dfe9","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/environment.json","sha256":"9b3e785e884ce97d61d89db710451a4a41f71782d6997c4c7183f1dc86000497","bytes":5341}},"requestSha256":"2b226db69c938aba7f7fcb3e466d1e10691a0b5551a3cf68876a2f64a0981271","metricsSha256":"543c355ff08d033865e0f5ba1fd4dbabda929fc13cb2ff319c8a3b56d2050a94","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2478578","recipe_identity":"9038d74f3bda179852dbddbd7b19e2b1dbf3baf8cdb77c401a89ec202f26a8e9","loss":3.639960767141114,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"90495e68c5868f7472ab2abfce24ce2227aa5ebe914c4ac2919a6701975646bf","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"3a6456ffb2589c344292796811c0a7feeca2aeb2c82595c440e1bd409dd38575","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"7b914bf5f732490e2e4c70773448d25efa471d0f83738f54d2b95912c1b15d16","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"0dc2c38a0e6ae85e116554a9bdb6e638e3dcc1508b53defe84466708dfcd097e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"178800507e030dc69302bf1f3cf21b2d21402742fe054a73e0c7d12ef76d9e81","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"6b9f9d93ee9a688e671a2bb3a05a0f50bce07789a2aa2986663adf9e143d3968","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"70e3f09ebf6cc239deeac2f97156ee48bed575112a4763fb4dc003a6b5179833","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"fae98ba412a59d363d0743212633b623b850e55afd6f64f6b9929935545e9fcb","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"c57324d7c22a7bb3fe15d2dea5186a48ed930245156700f6ca0beb9eabaafebf","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"b820fd734396a0b577595fba71879bd94b4081c4531bac478380e30e44fd217a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"0954a94bc20fd2e96856298c2cb00bd069b337889015bd993254d755fc066c30","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"99a02485c41d577ca88dbc89136ad136a6f150cd28203a986c3f9f72b6efc989","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"8894bd126cb78345483c19ee2e7e0809f55e6400a1e380761ddabe8365bd8ccd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"2c8f1e6e9bf4367d65e5025bf6969e9566ed269e35fdb1750ff20f9cd90b7ba1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"3a2399ef2b18e11651119fe16214b48b3ff724b46e9f2957c00b667f42c2160a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"f2d8818cdd953ae4c167730dadfa456e4cd5cbc2061bd1c1b5387019204987a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"9df57a395702404dd15634d3837ad7a240eab0ef11d756c3f73cd63bcc94ee0c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"1ed6b39d2faaf4753a53cd2c664dee47552eecb87d9326280374aa54b92619c8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"50c22d8f702e3d8d0fcd76e48a872386c7500c6817a2681ed827a56f8d6769e5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"eaf1e993b169912cee3ba381aea3e183437e97b9fd9d135da1d9df499cd62119","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/environment.json","sha256":"7fbc9e56ff5e5b9f7f30a9d6d6784bc8be8e42b6e168814f200fcb543d3d30b6","bytes":5341}},"requestSha256":"f0a09d43146986ef3b19a039b5fbd5500f654212d5666cdddad9d2e1eda4a9db","metricsSha256":"adff6b5a5dc98c2d4add770a8d9329eb2ec63e402aa76e60a1a8a396b9feb49b","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2477538","recipe_identity":"e49e7bce358bbb6d5a2c3bface61567b683447f5f5327383ae2d294544e4d9a3","loss":3.6576849139433754,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"4f398ba3a055a6675ed31209453978c9a2c29f6da6b04b725d33f9f586fd6b56","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"9c52e2d1d7f2a9fbfe7d2a97ecc546f7c4365939bda3419670a2d5647d1c2ff3","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"abc96e1e33bd5e50817812fa527a3e7177f6a930550f2e49ad58af1e85cbc77a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"14ea755f1d9c06825e811d297bfea42c68bcdb37b5ed7616dd4099c93c39c384","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"e7422b719f5a8b8d6508524b513dcf6390229c1d42048c7bb5704ecddee9aa9b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"f081979b274ba9991c30be03b94d077459e0ac1ea8d58263e85629ae9b1067bd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"26c10a8c39122c4876db3757611210da5044ba8c9b72cff220b35f29823b5467","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"33e92b03293bf73a3c9ec08fc742fe99283198ee3ec4519c04856aff71707289","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"4c20232df63a962cedf18944241de26ae4be3a73278cb57b41f9bc304a1f6198","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"6d597f3e4c7aee37565ab2cc59ab7f2bf59fe93e3ce5dd25dd5745c48a55ec9a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"18fe643de27c66f1ee94dffa2fdea7380977cf18b4bbe6c2057a9b046d7329e8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"c79a8279e9b7cf8472affc37562c718a514abf7792f9c0e821117ab378658958","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"657a38e519b132316b96014559c2421fbd84658461847b595cfac042b4a0f9b2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"659f2ad8036fd3aa5cdd4dd06eff8259a61c7cd89384c3f339280b57fe3ff91b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"6e2a7bfcb3c67ff62448322e338393dd06db2948c170e00d45dc33ca7a1786da","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"181db2a8546d09fa5cda5c8559d216c9557280d903cf9b802ede5560edb4fa24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"f45163e3876c72d178ae41349634c5df9bb04a2f11aeb63bc8db2b3db55ccbf1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"28fb6ff86fd81df4c60a2bf3bf163cd4cbc864d839e0a73cad5a7ee360e5f054","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"fbf829a366b8e3ec211ce2ec349bf048e3a4fc03806e444725132c9d4f89706f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"31ea69ee629d639cc6d750190412f477287600da9d7de44bf9fd6a6a149f39b3","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/environment.json","sha256":"20d95ccd6fafaef477c351244fd0059c1d365520d914a33d98cfd6bb08a731a1","bytes":5341}},"requestSha256":"ce5cfc08063d35f8dcf338930c15672dea5276b70bfea0d0cc23358f11ec093e","metricsSha256":"d1d4e797b97b3b3dd787a3130f6caff38b879c64000586802acb9db562353317","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2478579","recipe_identity":"41c3151fe8f5257a4a7760ea42c8e5ef5bd39fd081f9465c7211acf69ea8c54f","loss":3.6645932975147146,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"012f28f754ec2ba6b4df49a1882ba532fcc24e56e59e423d227079977f5299db","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"737e745b015f02347d8f8719044c8e20631c2b592cc2ada4262156e1759dc267","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"46c17f4254c4ca744f69726acf63f0af2e7e8129b190907999507398f2b053ea","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"c7a13b885999149c2b33c4e0543de1c7cd43fad138d4bee1946163cfe28cc1a3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"2d53e3da7d6d803681e1f07c8472b94f50b7f9d9d67c5ada448057955a9f7d49","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"a2349c63d01919cb6ba933aac33e39bd0bbb6b1678fb086b53ab90575c09c33c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"aadf2f18738c2824a8fb983f7df96583ec70362500750374fa74eed66b80222b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"7034b368a5fb5d2300d56476a0c3c532ec8f12f43c7ddc2bcc5953074ff1ba00","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"907dacada61be6b28547d97f31bc0de9ecc93e45685d428c2e6857d4bc71612e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"971bb2d1dcd8234fea8628d8dae88301d869eec30be949c63cebd8d661dffd2c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"5f5fd129c3c84696b875fc43a0f35e33e33b9433fc18aa251f81b37178d0e0fe","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"c16c3327eef088a36b775431f53ad54e114fe5a38b3bd3eb634f932d8c969812","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"3b4d6f9929c1996dabe9471f4c4905ca30e87d872d0089b98370659218236b80","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"6529d1668b467fa580c7962a7c699ec9cf516ca290f6fb229c84ef89acbfd7ea","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"8bab287813aa25d8c322eb92beafa1d1306fa32eb7d0b50722915001d1197a26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"503f7ee6cf3f67ef106751984bb43be8f4d430ece800926759acd7c7071a8a74","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"0b989c3ee62a1f65b8319855f2b72dd679e89122cd04d34776933bfcbe0c6161","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"6b402bbe43871acd9f7c70718c35ebb50a081f1d79773cd3e6738f93dfe2f0d3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"32bff50e4c52ba0395be8c4f6a7ec9553d175f352afb27b5d938d0ab17dc3b5a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"de9003901a19099e5af08b0bf83f9e00f4953e4b85fba790a5b45dd65d0dcd9c","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/environment.json","sha256":"8e7c3780b0ac860e3c56d88a231a727539a303f5e7598f24a2df4e23171825df","bytes":5341}},"requestSha256":"738ff7a4accfd16217c89f1b2c84c69a007d684b2c8c7b05ffbc84085b25c4a6","metricsSha256":"3edfa48518009e1b970386c87426122b36ef50e830e953a7610736bffd4166e8","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2478580","recipe_identity":"75e0318012134a782cf50c0e2b8345ff6ce93fd2c5f2131b614bbae2bdc872f1","loss":3.6504553762937646,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"a5b13cee2944135ef979dd37ad86ba1457691656460b231137fdcda46e1de3ad","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"d02c152777e523f7fada5897f01bc2773f20d9b32653353c9f207d6f3b0aca02","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"e2271319e40efe648686b3cb9b958261b58157a604fd8e19b85b142f398f7f21","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"e56808d4f3af12e013bec55609b12ada7779de8b2cf4c45730d7c22347383171","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"65099d3be4eaa832e29e9f87134955a20f5b79a3c5110508312c3b949bc5a709","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"a3284a361d8a55e9ee0fc86473191106b9c9181e9ad66dac81303af874bc61d8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"14c638a77c275d950ceebbb3c98502791834ab5dcdd82853844650c91bda25a6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"b9a2d1080f8849e1eb96d7d0d145cfeede13f7f6c20fdbac0614ad0bec44b231","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"3399ce211e728fda8171fc14bdc565589e05d3b3b4d9bc2ff67da023f03c1559","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"8f82ed31784953948c2f4df39e20903d2f4e1e314045a85ee44e9db0e050725b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"cc11c85391c15f2c95bf304b099b4a57ae35f17a9602760d52d02c31c33d1757","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"19615c27af98a18abd11b4d4f509593a9b703199368c6064d02988885027ac22","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"3cdb0feb4de85f837d30d3558269d0114a458ca15973373432ba48a086de60ad","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"453c0a895e05964a859922d2ac754b8d3f7ddefc207fe56285269facaf631e89","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"e4f079818866b3c3e346e9c62403bae9494753ee75497efd143597271a44ada8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"22018ce85a816aca5ee7197ca0db1b592e2bf4284b133d3a4a7a07e15efe77e4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"bd2d28ddf243e8f164db5c34e19a961499ee9de1eb5d4f5c63d4d313684fe0da","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"f5fec8c578765696dd6ee71ecbbd13fc8f093189debc510f067a3251bfa146f0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"c084a00898711ad4a21866fef48f604fa84cd4dd0ba4e91cd01c512008436e68","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"22e5ad642dc89a6b8b26d566be431a6dc93acddbaf8443fa7425362021249d90","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/environment.json","sha256":"8b7497d6fd5592c7835ec52fc01c93149ffd4f7a5d9fdc79d8aa54c8ee24904f","bytes":5255}},"requestSha256":"38271964e3627b9e6969a7ee0b6ad9c491d19ac321711243d26ee68648bdc2fa","metricsSha256":"c2db09a0c55886e9efb6253451df8559edf363ea58f0df4ba22205ab77e3e4cd","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2478581","recipe_identity":"c943918421b102d2e481525cb4c834f6cca701e138b10341dc89a1ef286c574c","loss":3.673098105839217,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"d92657714929a1068668307f6e79aef5106e3160388419c6eece259051a064bc","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"89807937051be07495047a1099270f9596a2b1812f44ce4b84ee1085ac3444de","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"236e1bf56be6e21fccc18332c6ba0e8be3d9bb4d40a3a0fde7a8de4d25715c82","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"d2fe375b80ead4aa95c81ee669377e87c095149083d5d88f8b425e96a570b8e9","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"4d156c4e9178de1cd57ae90409987374f3d2009b702717f80f3bf0108972d5bd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"271d389087b786bd67b6d5a50e0c689a7168409a2e12bacacb4d50116da4b428","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"bc1f77f1c0bb75a67ea050a634c968be32b2542f54936802fc213983d7c75d9b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"f6979cf5b004873c221a0da47c4f11f2c69743fe371c342b34eb367f47005a33","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"d91f85ddc24c08ec9d939e3daf3e3de12805f423cb5fa94dc1f659af62ae1fea","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"186a9f960111795dce4efc899a899790ecc40f4ffa0c82b6f83c1105aa775e9c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"854eb0417e466de69726a27b3c422cd0b97ad849a2546852dad5ffdfe24af8be","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"0cd6460a3e1b9fac6e2965eeedbe2713bfa4be6562fe3023554936e484cd4ca9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"5817fb136ef8f479704f9f6f85ff12dc5e7a9d92bc673f7cb05a03e38ba57500","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"465bfeb8fff892dcdacf05fa2696e98db9c1e88e052c26110ba5c8600ae5ae6e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"76fcffa2116fdb76a8daa16d5cbfce40efb8312eba88b15b4562534640f71cf4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"872112b5044b78986c9921108eb234f415ad703e89ff85bfe3e61c93b01853f2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"0c543e0441fd52930b0fce4583d482cca364de5c881fc8e43e968e84f2d3797c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"d8561eb37ab17e3eabb5a495ffed77e30e3a727e3333ddeaa0ec9ae681685a8c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"a5b72169e39a164bf96f821d219448454ee074995e74c3f9c400d314604283f1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"0d47725e472035505a43b7292cc677b61fc19e2a62ef67881f76fd754f82a4b3","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment.json","sha256":"c13c9f1b19beca7cf3a3285e181b3aa9dcd7c4203353c1abe02292e4b1750534","bytes":5312}},"requestSha256":"0e6a08d4b80f600c3788b7022d38cbcd4200679a1864047fb85f2c8d7b5bbd0e","metricsSha256":"247fbd27d616ef10267f4b7111a08a34538dee096e62b6c60562996e8a5b2b19","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2478582","recipe_identity":"89613778618199ab0e5b8bc0c6530a469115347253d7ba45b58deb8fe40fc7dd","loss":3.67417623649733,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"e2ecda62bf64449132b1b7291714b7717da794f53dd68d396f7318897ebcf497","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"6ad716083053fe554493a3ba7f811e321c6919e5fe5602bf2de1ad74bcb60a2d","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"86a14ab45c3c05f412ba59994275abebb8ca391d291f76ff196aeafba679f910","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"0922bbe11b992662c2080c14a436dd0dece417407c895d7acda92ed3fe83b813","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"6fc939769da70e1670b24ce4b1ff15bc461f089276f8020cd9159ce53a0deabd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"0d8e4ae99c5caf145600bb68e65674946219d92129b37f9b95eabdb8b57dcbf3","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"dba575ebe19cfac51a2848c95b77b007d4a391eb3b101bbe300bfb95f7d07449","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"947053d4f100efa323deca7912e3b5051895880c445bc7bf77dc186bdf0955f8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"5bcaeedcdc5df6f17decc6cf102ae021972f23004b0ebbb8ca40d46375a43e5f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"831c02153cb92de2523ca5f4adec20f8ae885c19b1a1fd791f94ac07604c2d07","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"62d97bd8c7d5891a88ba49ab6158170b7741c6a140d6329b42cc33e59bcfde6a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"60551d8921649c2d4153c886016f003307b4f26ee41bc73752b9c4e5859ef2c6","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"1c375b36852d0784d3c9b34306f7b93231455f77b8e041e29114eb02a0b0a97c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"630fdbfc3fc99a54cee2bfecef2a12a837d7f31fe3a4b38bbc3e3be247af6a92","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"843ab2b4a4b1f8cd9bf89b643577eed30d1ea742e639cdef5bc0d46189a1f677","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"ce45e5441aa785d06891330726cd2e062cfde7493a95ddf6734c9a23b8351dca","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"c4cd2c8b8ea4fb2974486774a2f87182739ce6abbc40ae0aa227b369cf017f77","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"75b09b266c03179e0e090cc1566ec0a88052b894ea7bfe11e792798e6c6f2b7b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"c4492d683260eda7c83d98bba903e718f98b267c5ba582e3aa7816fa91962c98","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"8ad1d9c1bb3c536597a32627e514c5a5dc5b286e5bdb4697e13ff38d9a04f748","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment.json","sha256":"0ac7adadabc4eaf6daa630a861f0df2f15cd2b6d7500eeac71a63ea7409f9cdf","bytes":5341}},"requestSha256":"f58a579cd0622a4e4e0a6e675b4188de67b281eaf67a58b8f428b1886cfa6d48","metricsSha256":"d5e72b04b98e1e939364e22e0283a89f2c96a09facbd6e3628a78c2ab9d60903","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2478583","recipe_identity":"d38853ed86b24cebcfbe30f227a1c3bb504a61f4d08c00883b7c0aaa9a77db6a","loss":3.671167278966557,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"e6a0a52d9643c2b98463bc8a9b4cb0f02c8a137f6e83b8b8a4d28b64743ddc13","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"233621a064a9ffe348e255435bab41f28c1695c914662e000addf2196f9d7d8e","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"9b8c1587be05a862d7488d0b30d10920961ee4a449893952af06221f0c8269fa","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"a1ba4cbc08cc55c2b8c49473aa9f637c767757c21baafe7d19383c3829d7781b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"50f479ce6e44d4b1f1672d1753d0754ce6aab871f1ca3f54d43bd66714135a68","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"2fc0684948d12dd46a60b4af1d719416831fa61cde90bada417727f0f44727a1","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"2c2824a80d06edd6252ec2ba2a245632b7356fe694753dab09c5ad33ebd1ab2d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"00bc8d09d5c2bb957daed42d113246bbf4b10396e3851d97986f13a5c0c956da","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"6993939f6705451b1f16def307d7692e9d7f7937e54642e10c2f98a8141f9a48","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"87d79cc55fdfe92811667291e6b87f768900af00bfdd600c47eb7cd463290de1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"1abeeee48c0173eb14c1f823b93dec4621291e8080696d597cf978041dc838af","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"fb2b141de2ea51be072d55af786341becb093df2f0dac6e25bc50c82c7fd988e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"07528ea65f5e7f48d10f4b8171434b24f544274208ddff82bb6926ec267ec979","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"c4c28654ae661042813d8d6a179a037ed111f756afbbe99893ec7407a9ea43dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"24e0ee822c10f725613b6481d7c825e7418874c7b425119d2a87503c136bea2a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"d29de7d0f5c0a0c71cb1a37660919b5e2533fc18a27e2ad6a377205f0ff1b9da","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"e95633ca94693e3e7a5a1c4f9b7f43fdd22dd3db31ca159ce07a704d3f37f457","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"19797c4d01d02bace6483ce0fece53efc9c6a43b45da8a12f495d7b3d4eb6e96","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a7a7473a82c9f0154e5a24740fea765f57259a9d2693e6bc8a0cad72b50c044d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"0b11ef69dd4d2269355b3f454cea991fa8a1bd35dfd703ecbf83069a1f2448e2","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment.json","sha256":"6ad489e42d6c8815002b38d50e10c53338f2a4a6be19a007ae2836656722b88b","bytes":5255}},"requestSha256":"51d156fce20ed795be5e4eb9e6d11c1d6045c6468bb905d52f70d8a25b94a7ff","metricsSha256":"6dda602c3a2ed6c388edbbb7f305094bc4e61077cd7187c5e1d4cd852ca593b9","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2478584","recipe_identity":"ed97b3688b0b8348cc06e2f8ab2e1541f97e102336525510a940e10b3c7b0865","loss":3.7095806459825638,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"5699cce87205f669c5d3ce2c9e8e17294594852d4ccd6c9e0a86b04d8f21c7a1","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"98e879024320528b1d6480f319427f06700d9ce3c352ea475a7f23fa4b55326b","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"a36d68d2330f138ee5147295b6af4a9be3a92f5a150296d34ef7faf160cb322f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"cb5b77d81da42d3f23b3deaa819b573949d8867ff4cfd8801f49fe0d0e87909e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"33b6fdf344c201d86f3d68599bc7ca43ad1a70e0a20be22b50622681c885f920","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"f7b88f21a477b3179d7da7967847e4f57778f0dddfc0adcd71b1c6a15cd51eb5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"09e8e0f661c14f88ecd5bdaee9f62b191c4816fc80b4ec17d2da200e4dbf60f8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"4746f3513a94d95d4387ab4713adc54f677161f552dd2238c009f80f51628ddb","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"f370e08162dc42f0429bc40bd0521816a4c650db1e532fc6a4c98e7c2e00c59f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"5bc0f2c04ae17a4bd9fde1ab5f10556164bc6b18b7cce49595f8d7ee4dfd80b6","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"087b2a0315a8d0879168df7cbfaf8968ac97830e330232c41a9ba3cde7c710e8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"013dac99d839f2a83b8d2ba48587e4d1cec56b7efdf4e36ca0e759f5c747fed2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"3b019786c7968617025da87510943bc4f67d80c9363a0d36fd8fdfbf83f0ad35","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"ec77008d8a1d30f7ce5b12285f735b1d1c1f80cf3140c356c3a9344e6e0687fe","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"db11ad37077a81fa08872b8d022f833c89ecd1237a9f923f89cac726e7bd8d53","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"c877757bc1ac71078ed13148827269a02037a0f0940ab1d4b1ef2ddc9999374f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"51242d462dbeeb325e5e99de64afcb951d9c40a08ef2f8eca54c8f80a4df27d1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"fc28584bd1c85f0832385ea3bff6b30958d243d528207fb3013f8303e80bf31e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"bcf71a9bdc7246f55a6311240ec2321fd7859a67f6b78690b3d47eca02d8fce8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"6e982e88b39f1f159dcd1d1e285997193b48192ab628a5b2e10d40f8a82ba0af","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment.json","sha256":"004c36bf879790b84ffa9ad073b9aa295382709369335f072a782412f11e8a8c","bytes":5341}},"requestSha256":"b836bac9902f86e344ce78d341041ae8d3cfd90e10e0b24ff7c6d35ea232284a","metricsSha256":"e05fa0f91b7fb043aca407d09926435b85a1ce8c82cba0a673e8ceb53ffec04d","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2478585","recipe_identity":"8f7b2a79acb36ded68e42cadf3663aa961233597d49bcf57a9a000d6d8f675a5","loss":3.692611871122898,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"1f2d84d3c80b2df8f3cbf970dad543020f7c4c521463c5a99bac80d2c7af5525","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"11b6adfe5f7391f0eee5c379ebbedb64907d20ad24ef3d2eb91a02db2051d88c","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"945b93c165035aab7d52b672a8e88f78fa9073fd5f7964940727f495ed056580","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"5d2fbc15e0780e254bddb1a61ec9d02117e0bd1f60fe316933d7bfae90a9ea58","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"d4d0921f91c3116b0117889d856a60a48f26dee0159a40d36948f8641bf52ab4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"3d8834e8165d7ee30cfecb9c8b7a98478c3bcc7c1bba217852735620c7c51dd3","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"d259980fdca91d75497bffc8d048f0cdc985aab8dd5efb56284c987bc2322f25","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"db86f4f6e0133270c5ffda99f8e271074b5278915e079de0667e0c2e7ddff8c7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"c24e02aaf5b917d5411f6c09639fd34aa5b39dbef217647cfbafc91ffb623821","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"d47b502b4ea9c461d3f94ed9c3ae1cd457610d95ed778c024a5faa3121545336","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"63eedbd3ed9aed3b5fc500531d75a3580885287955a1021f498e9266d7f74245","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"dd37d022d60e8aa9646518689096c9d64a435d013e2ec31d713b96c66f4d618e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"3396ade1bf20625ce132ad18778b0dbd0aa1d7101b98917c0c13176ab2c22631","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"781f6e067ca1ab072475dc1ea9a89676f202ddaa882bad6bc43fcaffd5a1e629","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"add80dcd1fd882e1d60064136cab7647066e7c40be839872725cf3cfdbed947c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"e97dee90e2c2d05af7d9dc688939c6c794302893fe70d1ebdaa56460dcc53da5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"aa33fb8275845ab2f42e948f38af1a999cb350c3fe1764d512f22cb5f77d9a73","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"d68aef98675992d957212f916083b1d324f34dcb3601fb317137e8df740d0fd2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"7a1a39261fbc65a801a49c94abed61d4dac57ccdd3b976bd691fa4cc9a55b5d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"b9ee46d6371275bab9a3f96966cff92597da9898dc9602767b6ca993e4ae5fd4","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment.json","sha256":"36ba10b70546f3333120d85c7813ed00fb0faaad737036752e509c26b2adfbd1","bytes":5311}},"requestSha256":"f39a9e42b6e03b8a8309969245937ac2cfc4ef7b11de40e91a0aa73c0d31bd87","metricsSha256":"5b8b4e974063f3d4732710a595649bf8bec50f4726150274cd2bfd5eaab2602c","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2478586","recipe_identity":"7de9da81981e85dc44d6f7055902dff02bd98c5427ac496120d6a20373635c05","loss":3.7079620559754964,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"096e6a986d3a3eb5b7ab2ae182252f41ece309f539b2d6b1878ce23dca68d3ff","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"7cdd1d2dda65c97d5d2d6b756ce976c7ab282469e2b317573badc4ea35e2e23e","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"4649f5a443f1ee84b4b0eeda46ac14dfaa6ba28bc503b546522bd85cebb62dfa","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"df77e8dbce35b9ab8e6ec17f1926a81466ae27ef07b3613aeb61184a53497b27","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"b0c57849554a7f7f007672d08055f03c7a2311a618019badc202a70cd4225e3e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"f6f52dd79be322ebaeb7615102d348c3d7860cb0974ceebf9fc8d14627ab8217","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"33f3d8a93d358069d82ba0f530377ffc854eb0f24acd7763aef45bb64f9851de","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"cbae0420096ddc68d980245896b3472573ef55a6d9661e0e3bd9ec0e91042724","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"3500f735dc151334e9fa8eb14d7fa6896da08338a87f479711c8f8ebae49fdb8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"999dc8e9a442f5e3baaa82a6cbe6e364458ad2108a392f59fbff0f7f0ffe3ce1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"1f5e24d174659da77a945d76b8f652976e631143cb8b28ac67c91fa8d6895be0","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"b5547b2915b29ea291c712a409d8508cdddb378d7c653bd14529d8834d9d3a1e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"e14979d462a41e8990dea1c49a11db23cf3750877f04b7efd5aaf3cf544631a7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"2da72d22c422dc08d627386fdcf8b89ab55a4a92ae2fce15d0f42bef4e50c1df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"637386bdacbd7bc13a5ad5fcda74bffb22fefb72f20bf3576d92146d9168586d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"61d44c0c1cfe2528eda20e59a115a6ef31d70490367429544463c8645b69e539","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"71e9811caa627eed9ba9002e5ffba9e4363316410bce3473fd71a49c5c5b8cc5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"47b683faf0d69fe8e7187c18dcc47736085abddf9807c2beddf6ffe325ff8e2c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"9ef01d11003a28154f86700edcd56eddf55de76aa6de31fc0be74c1a442121df","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"62fcf8c943197bd57b0d88eb97ccad606e7953a87ce89675cd9797f1c5ba345d","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment.json","sha256":"b235f342c1ccdd11e933bf636dbaeb8d67e71fa5a13d970f82782e271bcec311","bytes":5340}},"requestSha256":"124a741a1bc7d4ebdddf75ebb920fc98fd64f2cc005983afdcb46bdaa29e067b","metricsSha256":"deba764cad6bebcfed98687d212d603f93888020c0eb6933b3b292cd1a625025","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.01-s42","horizon":20,"attempt":1,"job_id":"2478587","recipe_identity":"d3d1561e4651e5b1c6592339f2497b4be88ba26702ea1a4d41876b7404d5c09c","loss":3.621028274060837,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/result.json","sha256":"54e437ba70e866eca9a1f438f3f9a3300784ab64856e57fc4bac62a82a1039b1","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/final.pt","sha256":"1dfddef46d9867841a7f2e3cb7bb7732df16c9f7ae94cc30f0e13e9b607ae785","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/final.safetensors","sha256":"8ec23fad85ca0cbaaf589c42770ac4971cf5f3fb9547184406102688e31fb8e7","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-000/final.safetensors","sha256":"20ed91b064d981484a00ba3003c786f156f431cdd0681f4d26cdcd027192efe6","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-001/final.safetensors","sha256":"842ba2a895d21331ebf5c544f0b80df9226dd12739fffdad20ed7c4167c7c1a4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-002/final.safetensors","sha256":"f22d84a95a064a6f213b90ac854dfb76e0a6f3fca515aae3796ace56559ee54e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-003/final.safetensors","sha256":"dece7a5ad54958b347bfd590768f19e4491d17ed960d999fcc4fdf31753ad026","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-004/final.safetensors","sha256":"3b86b6293fffc40d74fab3dfad3ee332138b84f686b2737dd547aefc041e2c61","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-005/final.safetensors","sha256":"bfe8ba24fa867f507e5a251ff01a25989d289d694d92c5ed2fe67eb6e43dd995","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-006/final.safetensors","sha256":"73bc96ad3336d8d21ebefac85895224af9af2453e462988ac0775e8b6e916a4f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-007/final.safetensors","sha256":"a78521f514fb370c3bd30d7e6194d6dbac0871f74853c63c3e7b5333fad0a27c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/epoch-035.pt","sha256":"2961a22aace9ff8971dacc7884cabe7bc683e49e66085069655132da889f76ce","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","sha256":"d701a6d359f3d3bb19633b7dad62ed5de94ecc29226ba178a9be15878b0dab9f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","sha256":"05d6e283ef2670b7c85fe12fecb1a6678405aee20a65ad78de9e83530a209030","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","sha256":"ce4e8420a4c61273bf771136d6fc8471f86d02d15132eb5df936f609d1033f9e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","sha256":"8e194a2ac254b347453da886665287cd37316cc4ba8705d8bd14683e48755e43","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","sha256":"d91e47776153de482b02f8ba58577817f60930f290b7fc4d9621abc8ee85a684","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","sha256":"143ed08a14ea1be9ca6f516f49f2bedb4f46749947374665b676180299ee307a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","sha256":"d05de73952b4150c8168f04615a6272a2ef6559da27fa1f639947faa29c1206e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","sha256":"597bc056042c2546c1bc470dad6fd8c0cef7b36c0fcba9b6af131a6d0a4f207a","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s42/attempt-001/environment.json","sha256":"fb66938489713f9f2bd8eec1dfd3c9849c6de96238f02432c5d7a5ff0580475a","bytes":5255}},"requestSha256":"6e447582d9d4d7171ece61bcdffd1961e5cfea4cea60ddc13336d11723a95b3f","metricsSha256":"034cffe5a71adb2642cb94f01b02e532ee92987a81561500cee7081ac84c7b5c","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.01-s43","horizon":20,"attempt":1,"job_id":"2478588","recipe_identity":"187fa2ae77a1a30f470c6147b8f039752ee73a8b9c446bcc096f010720fe10c9","loss":3.6182242029862475,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/result.json","sha256":"94fd004c2e2c044cd22707a8b3ba9583d87e1edcac416c1848c9d652a8b0aa38","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/final.pt","sha256":"0a07802e480d6b9340036c49043acab4bb94520061cb3c438e5f515bf9e4d39e","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/final.safetensors","sha256":"76cb2debaadc89445b354a696143496b9d11aa24fcc9a378a712850711bee2b6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-000/final.safetensors","sha256":"caa246cfe995ab7ddfa98ce3452c88578fe443a16ca53b05663f32986ae0129b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-001/final.safetensors","sha256":"71b50b3a832eae17ad6a677586c7eaa60f82fb14c1b1d5da5ab7004dbe0cd8f9","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-002/final.safetensors","sha256":"366a88d64ad771fb922c4cb67626dcb2324ac8324047fca21078d380f4cf4062","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-003/final.safetensors","sha256":"ebdfb39c728435989c9aa7a249030f48845a118d53441c0c69a0eb7a18e27bb2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-004/final.safetensors","sha256":"06a4ce07bd3a4e30d49146cb97e6c8c63acbee306a27a13ce92af8d885d10f10","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-005/final.safetensors","sha256":"71860968ac60e8554ea01ce54d2a9e7c35a2872c74baf0d64f91469d59250c45","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-006/final.safetensors","sha256":"f75c64f28475078fab273f9b4fadd0a45e9b02887de9a91b65fb3475a93ec2e4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-007/final.safetensors","sha256":"845cf82a435fa66c0188d839525695123051a16497696dc46f57cf9a040c53c8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/epoch-035.pt","sha256":"f95b0c8dd9a8aa879e1e5b78bbeaa6e5af6dada7db7263bf9880686d7b58937a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","sha256":"8d48c6223daa634c9b62a3683af2f6ab327630723cff007f48cad08eb9137606","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","sha256":"dce79672246f0013c5704f542a53d2fb07a8d47e011f55f0b07c3ed1eb8e0d68","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","sha256":"e44e1cdd85abca3ac92ec7c93d7cd8fc763c920e2ba58d7479b304f5120a712c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","sha256":"eb1066b0370c624cdbdc3a3312f5d4a30e4fccd6a960e9426e254ba1b7942e3a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","sha256":"c53f2edbef9ab539408dd3678d8acc4c94fa26bbfdba13b4f220c55ef4e44018","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","sha256":"de908a095579c05b06b7e287e51e7de5029899261f8d205732877a4ecac54343","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","sha256":"9c96e84168df2cbd9432179c2fc615962e6c97fd89a6b521f0f6af4833b5fb31","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","sha256":"2bab69fb8511abe2685cb101b9bddbc452864099d0a6a8302397055fd99884ab","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s43/attempt-001/environment.json","sha256":"26efd31efbfbe0cf1f05420e375062342ce7b551f427812dbd69cf63df8003af","bytes":5253}},"requestSha256":"975920d1ad292eaff367921594d3aa5619242939cff909c85212af3913b97a87","metricsSha256":"8e1ede5bbb88946e7eac0d24d45f2e3b6603f9c693a832381b4412fda326c061","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.01-s44","horizon":20,"attempt":1,"job_id":"2478589","recipe_identity":"1c1a5a2e56ae968c92d36ebf042ef59d6190527c806acd0958abe7fa088206ae","loss":3.6197059886215275,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/result.json","sha256":"d79b9462fe00a934c60e1f08a520661cb9db6a0ee5162c8f6aae7882a38c787c","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/final.pt","sha256":"606b1a282e3389f05a7389176345c675a167fce1e271c5e3a9af361cc2564843","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/final.safetensors","sha256":"07b38a730ca6514178f78007be90723f294535b8e30baf59d846731c772d910c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-000/final.safetensors","sha256":"da30985798840533cc2bf8ac1761490b5f7730ed011268a47d8a313b95ce6f35","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-001/final.safetensors","sha256":"0cc5c85ccf29115792362e1bcba2740a4c93e91a6f41b329aab4e21c5bd2e8d0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-002/final.safetensors","sha256":"2877c566fb0ae9a1f2d7cf541069e4578a41c63981ec935ee3cb4e8d67b16cfc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-003/final.safetensors","sha256":"d6e2c8fda015f40fcad2b108cc84913915ecb3ed71145c7067b1fea1c04a13f9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-004/final.safetensors","sha256":"9490aac061257d9ec902266f2acc862f31726f05b23709bf7867c777c4bf46ba","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-005/final.safetensors","sha256":"55dee80dfc266da51cf09a3cfa3137dee2b7f15f23b2183692f09c10cfecabef","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-006/final.safetensors","sha256":"2ba538aa2dfc6d5bf755cb0125baec4307585064565c8543065378ca0879dec1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-007/final.safetensors","sha256":"b1d6f3df7ca8ec1c73904400d980e4e958a790b13c3f1677e5ad22178b3244f0","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/epoch-035.pt","sha256":"b2788c67a96c2e651779e303f97dfd074cc86cb9f73ba1c9d078a77b3865826f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","sha256":"2df268dd711ead819d7f11d05dfb331c9af152f3c3ed2bdf2050160e7958bad9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","sha256":"f55f3f68123a6126495f1ee75b491cc22db00a5a3fd2065fb7a551ae8ca03eb7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","sha256":"032b3c52c178fb384c9708c70bc77986f0040b050332ad992341ea28bb661f49","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","sha256":"8dcf67c5f18a7554c4836eca00de5c6dd2062957a40aa9a0a7fe9661b560185b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","sha256":"f0350e13c37be00e946a2a500d3618e3c633b0192ca4936eebdb993a308d3d8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","sha256":"2153eec21cc9cc14b2211a1c293dccf7cf5971bc1f94a52722f80b91b357b252","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","sha256":"74fbeffa08c2ff717ad23ddc031229e6ca01aa353e7c6b3e85addce20925d92c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","sha256":"480850bdad214f55622966e2cff0a1a36b22ae17614507b2bc272a07b686b7c8","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.01-s44/attempt-001/environment.json","sha256":"7c39787b67620d80f33418a965ebbf14b4a787d6df0a7d8d329adfe9a7ebea47","bytes":5339}},"requestSha256":"fcb66630d66e5fcc92072d2c5bf4cd9c8d0593bd2bf3453a946832e3e4811a0e","metricsSha256":"96505207c57c72bb73cf03d87aef28ed29e624d761bd9442cffddeb223492593","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2478590","recipe_identity":"921f1d5570709f41dc23f8b19fa1906222b3d15fdba0f02bdae395df286a4aed","loss":3.619966783967022,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/result.json","sha256":"1a158531ab8d74420277b8aa42282e410abd0e2b9cf8e525570b072a179e46bb","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/final.pt","sha256":"677a32e0603546379d511fd5b7d872e42d9ff2b1f85ad143e75d4c3158c23164","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/final.safetensors","sha256":"36c755dfe0c8c53f7cb80bbaf65bffe77603fca40162ffd98b1d690c3ec7067e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-000/final.safetensors","sha256":"e97ee298f447b310307b84d6056daa6273d3d48a05ac8813aec0a216501a9a56","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-001/final.safetensors","sha256":"74ed799f5d92ccfc4701693e04737f6b479f36edcf5a2846e4cc53d629b434ff","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-002/final.safetensors","sha256":"cc1f60e24140baead134a258364d012366ee150cbd1c677cdc69d63377252617","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-003/final.safetensors","sha256":"30f5370d6a84fa7642b4423d5c814d358b11d5b6e4e80e6bba3e31591bbd77d9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-004/final.safetensors","sha256":"b7822f1b2db6c417daa26c4fbccb9010fde2de92c0779dd9a74049a663b349af","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-005/final.safetensors","sha256":"bfd221e10bba9f142d9e026dcf22d69af4054cc0d025252995b24d2a88819ccd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-006/final.safetensors","sha256":"201194dfa48f9a23e9280e8023af2c1694987801a1df17a03b158224feb54fd0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-007/final.safetensors","sha256":"697750c31d6c5c875be7421a08abc942ffa2696263b5ba7cd3b7c4b7bc025748","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"38124d724a3b1a4102a629c8d8b19f3a6fec5c94c333e18dc872c94a688e9d2f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","sha256":"72025b58b903dc6f8acace19453e1bed171e426b93388511522cc60c9faa801a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","sha256":"4bccab569c819295cd88180138452d5413ba5378754235280f85ad11826dc941","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","sha256":"8a2d5e3331847c9a754be598c8f90c011ebabc313036e9321af3a9ea8325b14e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","sha256":"25f76c798d4799547249a13823d4b040cef31b196cded813949e55a0e3a5a088","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","sha256":"9026d21ea7446460390003e0c5750da77d9dd917b408da59545deefa2dbf15c9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","sha256":"24af3ce34e6c4c6f41ee14137523b087d4fd320b64a4e72c87a9ad07b90f1c26","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","sha256":"8311c68e479f5c16bed0ed29e182f1d6caca52e034c040fa7bfe9388f0e88eb5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","sha256":"3f4142279900ba1dd495afc5f4e7d541f42f6b443e70154973f57c93bfc7b7ad","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s42/attempt-001/environment.json","sha256":"d497ac53155de0386255e8a35785b3fdbb0199889a00e8f9debb68f4dd69a466","bytes":5255}},"requestSha256":"0d6d26657b4ed51939e79ab05f5846b0e74c538cad1465f938f665393b5ec72d","metricsSha256":"bc9abe1bb33bd38112e466e3110ec5ac79c36d03e88484caa541cb90ad3f5bd8","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2478591","recipe_identity":"a9f36b087ac3ee9099e4c89c4e76080c25ff1a1f064d873e9ac0414987b62a57","loss":3.613442201490631,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/result.json","sha256":"4550cc039ed27cf520a1cc17828daa01bf01b412e0f0b5bc126482092face8d9","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/final.pt","sha256":"5bccc26305203f49f728fc98c65b06cf6c48ac6db9fd3fad277056e07399a5d5","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/final.safetensors","sha256":"033e1e0038c8a5eefe421eee18f4dd35be86536c1f6e3ac6fbbf578d9157c5bc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-000/final.safetensors","sha256":"40175380fccfa66812546e4ed57044c7d0c87a9100a4a4bfd6cc38e236fd0b9f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-001/final.safetensors","sha256":"f3d43b79ea62c32c7c1ae4b25a082ae6759177ae034a3b45ed68b177f7db67f8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-002/final.safetensors","sha256":"6894c13b72ee2abcad66e3ee31e9a8c98105f82c8109b236904291ddc295033f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-003/final.safetensors","sha256":"ba5098802780f54ede6fda074bf8610e8a5ad8c8ccc3ee9f2dd556c215eca549","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-004/final.safetensors","sha256":"a28b844b869b4153e02261c00e815a03ce70f6d9d507ea26b43085dee604a72f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-005/final.safetensors","sha256":"13a2eb53071baa4476a036fd1d4992d88b01c3399d10c980adea3eb439d50381","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-006/final.safetensors","sha256":"79736de467af1eab03142507cefff85d7a18f4e1947dd5553041a4a8bba165f2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-007/final.safetensors","sha256":"97d85a9f7d39b6e0c18dd29a49241b59776175cea9e27070f0d372c803163a8f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"3196f12e396159131a36b1ca5e5795f5a6b7cc9006b3d1c08c0eea598721a5ea","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","sha256":"476a49ef81266fdf040affc54f1168969da9312818acaa28ca566702cc1cfe5d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","sha256":"91adfcb8dda531bc9bf738029ba4306df6d7a4f7ddfa001f4a03be7a0b6e5f84","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","sha256":"7b24397a495cf563fbcc2b7cb83a01f5ec6a3a7c909c202a872f7f0820bf9e94","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","sha256":"a07487e2236a12184d77f50fc569c822ad392b0229238396a7f2fff483b2a76d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","sha256":"28539f1d2da2ee6bb8ad151cf5771dd7b60514eb37369640f451d4add961bd6c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","sha256":"07159d2dd6a429986a2ba7fc86834b55b6d57f2f43110c85911aa8cfe9629156","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","sha256":"c08c2cd2799990da4ca27373c723111040ab39ea6e4c129a394e9f97e31ed9f3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","sha256":"dd3e48a134a94307177ff341e79132834e28cc5186ecb151212fbd8f4a2cea96","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s43/attempt-001/environment.json","sha256":"b7c79d2e7d2fa175823a536e782b62dba561a490c749954e77a33865b7b741ca","bytes":5312}},"requestSha256":"717a1cacc4f20feef5f0e823fadfd35ba4bac6512dbc8c2221dc69b456fa24e2","metricsSha256":"3c7b9f62d2b4095d3971d006bb552537fc60224522cdc9d0e88eeceafbe82112","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2478592","recipe_identity":"c2222749be73cb548d71bc8e075a7ca8cc018ef4945520d1f44bcefc093d040e","loss":3.6279922378386114,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/result.json","sha256":"c947a7c535ff905bc9e7c08ba146b7d165cc1bba1e4227b17200b6a01432ab41","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/final.pt","sha256":"f128e1f6ba19c4e369f18f5fb6171faf2ee0e1c567337488dc258b0b43b68638","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/final.safetensors","sha256":"1d191f0bb0dbb487ddf9690a8e3f517137078b79118630cf58f607ea88528814","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-000/final.safetensors","sha256":"590a7fb41f9ab22838dc875048c173421449bade29a1268ab1fdf311e83500fc","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-001/final.safetensors","sha256":"292a8c7ec474002dab76acf7610b7b0c4f90d609f06d059139b6c7c9fb39cb9e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-002/final.safetensors","sha256":"87a1adc6c0d0f4ab8ed3133f863fafcab5c1edd7a70256278f0f9f013cd412dc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-003/final.safetensors","sha256":"720554dae503dd6216538f2c0aa0a93ab46731b1d2dac443dbafdb154aa6cbe5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-004/final.safetensors","sha256":"1eb9559f0a260f0b49c8ad7577cff3f2147e587cb34d63fe6adf371153289682","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-005/final.safetensors","sha256":"6d6f78ff0adb2e7e85f8ce9dbdee6b581461722db563bf21a574c0c4ba319fa7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-006/final.safetensors","sha256":"9582d03ea0ba4b1c6db594512ac3a58eea3bf0b320fbdb4e27d7359212f899b6","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-007/final.safetensors","sha256":"1b07ac375accefb29ffa352272174dccbe85e3bdbcc8beb12e711bc4677770e4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"f06af8b7c2e449be45cb47d0584b3e6aac7f7196afa93334ac1fca31a0ceaf1f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","sha256":"d189f9ae21ea2ff37527f11874974f668161d9a69433886eadb94185b6bd3419","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","sha256":"794fdd58850601720faedb276eebf826e0b949ace47cb5b43a5981b29855a2a6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","sha256":"66baf69b536d55d33b8a56e9f373d81c7193c5d309818e5bafbf9beb74358bf6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","sha256":"f7fecc575bb1a022d176a1059eb5388b970f9891f450899e111323f4a3c114e7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","sha256":"3794f46a97a36e7eb7b1f093f72cb6845f3647d81792268e1da73d10a8c71b69","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","sha256":"31963cf5b58d2d0dd9f806cd113efe7fa4da93a9b55a1b5425d0c5dc7030ffb0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","sha256":"8823884443acc1c5b0f9efeab4a2a7574210a115dccacb3a2defca2ebd94d360","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","sha256":"2ef57d9ec5b504c55b549e9d2ae2053ddcac803cf2f7fea5633abc8703cd2a87","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.008-s44/attempt-001/environment.json","sha256":"0fd17e58e17b5c37c13764ec5b08445da611f7691481c335305b504400d92666","bytes":5341}},"requestSha256":"d49d7f0397ff4bdf30bc543bd163ec4235f8f94e225cd257c86c4299e6b21c8f","metricsSha256":"60a2a303bda11407a1f2777745fa8e2d176c284fe56b57cf7e07785252f78bab","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2478593","recipe_identity":"06ad20af0ff3ffbb2fd364f492421711810cc1a4f80afaadb39c7e2cf3889aee","loss":3.6054565734078152,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"d051e17f7953f968ae899d3adef10b2474992b2f4242b1a95fd0229b99b4fda6","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"b7c0857e54264d5554d110b91db2b2cf2f439d7dcaac6699da1fd77ca246849d","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"7ce87243cc59d2db650f958245bab2fc6be491446524699f3f24bbc368aabdfa","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"7e89c708f3edc48733d9acc57d6f109588f27ab42f18280f51acaca2e79bb3ec","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"e11387f334fe9506fef458928693d4e6eb3db284e7d1792c5f4dea7b5b00b0dd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"796d63f7417ae1139d61e3ba7307b8060cfd1a1e9271025a7b4259b76f58149a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"84548a43393323011f2b2f8624cbae3d9bf0c829000236a076ba9b463824dc4a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"190d552c0ea9439969002392228961b4c33210ad78aebbf20331f5152ed18220","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"343762659039c3a5b4fdafcba004a36bb55d4f64678422adda71ffb05d779e87","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"436f08c0e02b8a7645fbc06ad22a910235e51577e548a3d6618735b6ec7e42d1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"5584519def67293d1bde2b0494bd0761dfcec1d80a031c37fc5731002374d1ca","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"a48c4be34b88afda637ac911b6265e318d29e372c81bf64953c370a7acfbff44","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"9efaa605c71e410e1f0dfa527f11b382c38bb35dffa09bb2ce5d62f19e4d5f4d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"8be8f54addc7d6dd78b19dfac4fc94c66c572aec41fb7b63479766166a8fea68","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"50dd31881e94ada00701080e787048a99648f65786e512d154a27a884c15b1ff","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"4dd7f5ab12da0c4196b6c8a4b2a9513881eaf67315b949b6e4055cda29ee89db","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"ffcbe7614864f23db6f7f17091af0a49f5bbf42ee662ebcbf2d719c625944c66","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"78056e7a79c02d938e6d4904ca51203419e4c02e5580bdf81524b31238f98ff1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"85f17bebdc323fbd938f42b39e61f18fc59e1a41dc7146b88e25e5557a9c0537","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"96eb8e30a5da1f0557903c94646ad59879bd7bb47ffd421ddf914c83996461f2","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/environment.json","sha256":"f7b58c2b5d06949d6fe682bc691a170439c09bb1eaa8c063815d021cd0cca4a8","bytes":5253}},"requestSha256":"ee1d8ec42601e72e0b6c40e2792e005ea073c9e4c873998be93b4e9c53c549e5","metricsSha256":"6f912379bb5b84f48b7c0f38671ff519954c326180f1c862e907ed6a35b963cf","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2478594","recipe_identity":"5723a0bfb8869dc896b9c9e3e8d79ab87ae77f24d4df9dec52f5f4c31dd2ae50","loss":3.6113107910124556,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"09076cd754e3523f4950ca106e51ee49c297eb1c506c46d0461de7c6b7635285","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"c2fd368c94edac6ac7aa1931a5b30ff30ffa02d47757af3a1dc9726f4eba3e2c","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"7249cafd579ed6804ec0a538d9d2489ba1f758f01ab8923904340f881f8d2a9c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"8c33df351628ce5360cdeda2c4d699a102af1a377294932218586c92508460c2","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"e9732d66a091c28717495ce75ccf2557418de9ce4f36c6a9a00dd4353638d609","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"38b33a6957711713c4c2c41dd48228bfc261edcd84a5c4b8e736b411c760609a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"7a9c2cf205544a106fc28293761f3c1252b10926638abcdb83b274e29ebad7e8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"0b72f2e149b6f3d2d0d00ffc9905e4cff5feadb2fe37932d1d90b16e93040ebb","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"77ab77cbc06ebe00c0ff2e8f17c4f7b56e8321d71a77396f27697268c3d0a53d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"e7b8599cee597de88e41c505ec0b8d7d99fd183bfe66bd762a0bd7a5402d06c7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"d8e6a850e6f80e2c9eb3d1d3ad7ece3158cf362b8af7aa3e227c757fab460799","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"f0baef0de4cda403b45454cbe33d23577bacc382a4f499c0c7a9e3c0cf5eef8e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"34fc00f4f2825c721b661067fb2b9164fdb3032535ef5eb32f567052a6a4f445","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"7b7f22ce340e26f299f71eb707acf6d79d987827392e7821dd387e3a457c8c7c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"cd72b7b7f4fcb93d5acf93d0550ba03ce39563c06e8c912e9a6cb5eb84acd258","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"1651107d6328e89a5fec8f9e3ee7377684d278a9e6b204d96236a1f75068e774","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"521e7b042a7fea94c90d6f4e38ead101d00a0a4e334535eaa77f2cdc2c862a5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"a6a6ce58190ea51e7d82ced6de2cc9f0b580544517bb604cd29b66e0ccc1498b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"be62b8699c11b27a1ecc2e410cadc097de700d556a74d6cd197b18d4ede38fcb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"c3c7d7e5f010e28f61240e311d0198229e4fff7f3790e7552b3b0032d646d052","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/environment.json","sha256":"a5625bba53fc8075aa549741586854421f75fc5c99393e2b19df43fefdf7fa5a","bytes":5310}},"requestSha256":"193b710cbd54f1cb8cc23c8fc3a662009382bd8dccd71965db874a67c8ec6d4d","metricsSha256":"1f36bdd99ee7bf429a42e375ef573995286e7b38cd53c34eda76d0117b4919c5","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2478595","recipe_identity":"68caec269fa82ef5c07ebcc94335bfa05e89c7cd8a5e658a7dc35dd0f4ec9df1","loss":3.608697753371398,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"fc1af649df7adb55a661c22d51d31ecfc8610f528f2a3832e740a5210086c6a8","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"3858ef677a6f25551049b93eae21f1fd60cae0181c087b8314ce9d9304979c4e","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"ff4abd6e914857fdd36bf60d608221e277fb0ef08adeae14f9ad88bac06a53db","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"ae6ea659c351efd4cd6151a996375e4c746aa2bf10eb7b98e5b8e4e285dff2ba","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"6a83cc62d371e77338a287282642cca57bd80e0339dfafcc83602af333dfa41a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"6f74e73721de822db3b1883e0bcd768dc23a63f8286bb6f942360e78684a6fe0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"e34702c669fa760c20f8c7be865a4c63ae29ed1b65dde64906e0c8fccd7fca25","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"cb3b17285c46c3399d92287c12c0ca210c0495cc168046fe7354533b5a77f4dc","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"fb6ab286064ae19ec22053ae5dce2418fbe0fd6b649b17ff775a2154e9354c97","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"d51d959939f65f177dc15c2eac1fa7091d6977ea31cf9227bf9dfb4656a75bcc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"3e9b8128a103eeb942ee0e904c822a846d598ffea52f481a207e5018c48c7fb0","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"f70e624f5d4ac6dfc889004af2761d13a0d1a2c355dd595e065a0ca6e30e35d7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"394736e8c05552b6c187b7a41340e54d9d87745a5f3603108ef62a03ed4590b6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"c583fb94e8dba7b121d0be0ebd638c11c3ae00e5c4d4b0597971ef6e32b59a14","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"3014f00ffc83d7043d7ba468b999cee7eedb00c81ec80f5d49e376d8c6ae34ec","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"41017981457cb5316dc2eb5687e1c5dbd3f375c15b3b3208eaa0ffa3c3873ceb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"f09361233f1347836e78923f7b1777b87dac9fb4f6bb9caa330dfcfd932800d5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"94859e1acb165260b66630b156a2dbc9fd6c0fcff10b6e337d41d1550c96e415","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"1b05e519d2029806f14e9d924d0730b8abc2e833d666b7af137808c46cf78ee7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"cbf7b02634f2e5ba60c2e5172da033f265fcada45af3608e9086c6cfbd596034","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/environment.json","sha256":"6dfaebec695342d0817cbaed9327e84388c02d8540c3b2caac258eb39b574b1f","bytes":5339}},"requestSha256":"5d48f10dd4a7dbdccb128d2725fb4c78ee68f03fd38725508b2dbbb9fefc3f55","metricsSha256":"9d0513b3a8f4b7e345a0b0dd2b190e9c7c4932e6c10504206c06b22d3f85caab","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2478596","recipe_identity":"781180035e2e2711e98ce9d06f866738d71de2f60fe24d580cf05e3e1d7e42af","loss":3.6178593849568252,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"7e1050fb0163e02ea65edcc7a8e1521cf1baa9b5d1e7388ad019c43620ab2528","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"c94b9b97576beabbbbc53de81ea7b81dd2971bfc47132e17f99a7982a67c0c25","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"4a8fa270c6edb16413f1f3daca7c48cf7e8c544e15af73a8f38525addb60c256","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"6fa29dbb757b7cb4464b48afa9a3a44ba2cc5f4f416ec8c997039a7edbc8cf0a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"1d32bfd840e0759f6e8afd30caa932f677c83a771951784336179c9104da4baf","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"eed13ed053f409b9d2bdf74dc9239026860d8a6fac7510929154ca30c96211a2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"dc14e14eb0d501f585a171193ad10a0df66f667d12bdae504bf44f685e06ea1c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"b33a2220bff1843248e3e2d1c19c8765b3ac0aea12a7bea98f0b2fc44e6c92d1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"60eca5efbd9d2658a1c0624ae3391f4f9ee7b851a0f80dcab5c664878d5dcf3b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"65ed619e8f8b6088a95422a0cc8e8ace25c2274cf44fa4a1e849b14add923566","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"b6cdd2e8f3968b6e1bf7bb28318d63a2b12f40615b69041e545d86d4e089d450","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"da1db943fa1d626cfcd3c819037243248740f9f76be7dca0129d75537a35d4c7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"801ed5e36168e7d871dc429bc3cbb0d434e73eb1197e884f9ab4d50aa5fa5ff1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"1ecf69b793cf1014f14bc2d90398b121c7ca383425256a2cfad346d3ad58514c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"75822046ac41a763a101486177d9e430e570c97265d3edb81138eb43e6735a27","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"0e6828a08ecc165f86d3c5a82b1a31179ad5426954e4a5d92a6e31052e0166a8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"3035ff364ba75515d773f17f151f05058ef07113b55a6c8a73b0d5e10ed5c9c9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"99eb69a532e3491aa50d8c8abeaacf1eacd5423b9b5adee3d7df2bca1243351e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"5aaee6ff692df9b5e4cfad6f5e765b82cab4ac68488011d3f2dd0acb88cdf299","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"6ef178626deb317e7a9f0a14d6011d7cb63eb0f1e4f668cbc1843f5e4cb7e1f6","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/environment.json","sha256":"ebacf181523f3f9ff799868f5e5ed7c9f3deec777e329eb5daa297006b6f5068","bytes":5251}},"requestSha256":"d1566f0f49f4ae8e322fbcbf7665bb72292c88cbc53400d2df93bf0de05a5c6f","metricsSha256":"3025b6d8dc91843058e73225aac3a016928151072ed4931a8e266ef1334ddff0","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2478597","recipe_identity":"e479db68ff6798c28fa3580719e409bf05295d69a39f209bdc4662970ec8ad3e","loss":3.6142982909380508,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"47f1b39849d20ac321e937994ffc5fee131d33b984ac2ee8c99002e9fe88d697","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"3ab7b2917f6cbdeff050c9356c90b0c780f276986007f08f0372d8dd9dff9756","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"8e91df4d41eea042503c240d774a275b467fdcc11edda9b6b20ad99faca708c4","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"eb048dd0b319f5d05e9019416cebbe30499655947de22cc310016a9d6d7d3df4","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"2be2052fa8a77c8dac31bfb98a28699c1167ee8e803eb6f2cc0b24fa922313e2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"4f71a91414315d25e216b1e61885a4f086d37de46ce06b90f544f849f1ac5967","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"6523a3ea3e44e415636e9113af827a8930a58c949a9e72645fc1fe6f76d117ff","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"b8f56551d3239dadf1135c5bed8cad6c9ad47dde2f9c4f53da215f2a4c620a24","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"e7755bb6cf0aed47c7f31e4de30436b44fb31e84a2d33da81c3c02ba0972f2ae","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"6ef6cd877b7cffb8f209fe1e44d04fa336867c1b5c92457f1081a692ca5d36c5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"950074b3301cd3e1a8ad5e6aefe0eb76f6a638cece3dcd5e08be508f81627afa","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"aab6be5a39ddff107d8b0d11220a2274d10f80556081afec3d86e0188bb5ba06","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"f32313af8f8a8564ab9457a85848cacc694418aa05524ffe10031a70e5cfc981","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"6f107bddc7cf5df6d5dd04238aa89b1228576bcef76d67d7924bb1a0e203e7ff","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"0ce43fff10632e4591f4bed483d9e25ccce62fcaf388ff9c7e95a3bdc0c54115","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"2c3df5ed1fbf3ae6204cdbd926bedce47db4688a89781180dbdcd3a3aff7077d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"fa4557757d82c726f0c8418cb3df5233deeb6696d9550eda6c8c74864a4cfa9f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"02e2ed6b2feb73dd4d969a98739ecaa0fc78976f1c4bd0aec8da1e3bfe522e8b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"5468d3d71664e3c29c4421668c4a6104d56c70494766c08b7f4b34d2846dacd8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"16ddba3a0184409307007f36110b1b01c25cd9e1b42172283693f5537552277b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/environment.json","sha256":"6e0e0de5a1e7b81113ceaa2ac65a6b4b425e09e6e2cd61a9842763b9197e40fa","bytes":5308}},"requestSha256":"9ef1d8fc4edd1bac98f10b5c54e0e31aa3b504abf005ff26ec9fef692d8743de","metricsSha256":"1bac8e7fd0e905c866c88778a2350ec7e4b68f6e73b9b168c36a5dc7bcce63a2","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2478598","recipe_identity":"2e0db0f9e70fd9825b791ee2c896c70483b784534d7244e5bde733968d3f8aa0","loss":3.6211837708759087,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"f7df2b946b0a32785afbd3d44c5d57a99b685a89129b7eb92b98218fdf6ef2cd","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"7387978567ca0d5a168240494450d3d58809262a181e26f281e4d82e0ad7b8f2","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"8165db080e610f6c7e37d5f27223088cd1598bac5b3589902366cab05e57b5c8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"241eb64164e9e79bbb9386a300060f27adfc0834a6f743358d27e409321f7912","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"80815d2a3b17ccff6f2be95ef0a8bd90166cc7fe931f71d8fe738e0bcd86dd9b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"d8da833279d4ff70d8a1be14c28aeba8f8548e92d7c8c2deb2a295ab781d5f0f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"224d7d6a1a3937dd33d02e25604bee39af954e172dfbef349c57d845480629df","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"39c1def5addeb54139764fd842a91ca16cb5b566dd4ab456b9bced8778b5563e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"c46bd77b9249bc862b30ab3e8d9bb84956bdf17747318d3dfda9e6d7795ddc8f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"7186b807dd2493a9ec4c9f501afb7ef40fd854acdd211adb4435a954909f805a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"2c78053a28467f708260a3d0109a583a7209a7e4d1d40ac974fe85b873e39c5f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"81228b4b78309375e3956815d0b4aa87a747a1cccf94475322132b3b2c2f1f8b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"e8bb668f0c36823029bf4a1f04218032cecace0e9047e0170f9ea4291c26626a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"dd13de6bf8d08a263de8834c7391f5ecc19943a7f5642859deb5e38459e2d9d9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"13761934223b5f9d675a6738857254f348617793554712eda395e25701540b8d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"4e47a983a55f424c6c389168ac212b16921f9b3d70c497bdcd44808c5c2999f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"f1d1adfc88d70cf92ef07f768785f169b7b927996f65930b56d8dff35ee7fdd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"de03cc031ad2b8a230fc439acb9e2bb5076c7d43997f14d3b4c1f654ee5c4153","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"9d8e94d8544933b7f925706b79bd92adb7b84995bd595f196404f6a0848266c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"460dac899953bc648e56684d4cf05e0ead2fe0a8536a33cb8c007cf7a9e709bc","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/environment.json","sha256":"cee3f956f0f49d0bd38b7d51bb6f3ade7f7d9bb5dac247dd75df799e4abd9648","bytes":5253}},"requestSha256":"e751ae10f2e9fde34be9d4f2642464172dd4ba4d7bf988e8ee12605625d8efd7","metricsSha256":"6427d5ad9cb58dac77d7902df4b4f1dafbcb0d1dbe9028cd9ca17adfc0c477a1","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2479238","recipe_identity":"28f0a2c5daa3a77c6456d0dc46c63b18f45ec3682f820bf7cb18235c839089bf","loss":3.6250492355354593,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"8d5e57af967889f5c626d90814529f747ff089f0dc02188c37933afc5c0e91b0","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"72a86ea3e27b367e41b30f3fc61a02c55f44ae936c8f8d999470f15135d34e58","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"0cf6fe1cc739879492efbfde58c1a7c71cd909ad3f6016bc4826fbb927e99408","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"d55cdd209243d9485bd78714934fa95291127c84ee4942e02613be96a672eae2","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"38b3b76e849b65fc778f1431a7924bc1b56e74f8dec8fe37fad1a1874e7fb4ac","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"bb4309a0aa21ebdd72f0680d093d28644494267d4e3941e148c8993acadd6b3e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"fe96d68ff6e972c187e8db9b618509fa13c3fbe7fbed544230f2c94780e0ca16","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"dfb103366413be4f16b7f2f0a0af7c79f66e76f92f1fe76557fa9ded514726d9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"054193e535c5921fbdc5900f166a8a36c0128c642aefe07e7dbfda10cfbf7380","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"396a32823394b0b119c82c20215483d979eeb715a9581083c1195bd80fcc877c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"4bba52a12e88b2b3d624f111108874931c1a0237f3932276a7835fc5eedf6a22","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"dba64f0a1eca9f334dd02a53f6ce76278c6f19de1fab6d115ad227f1cef62622","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"d705e0c50621c65fb3a018118eca92602e5f9dd0b70fd18f86a4625ad66ccc0a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"1213969a3dad5befd45968fc27c9ef81cffef0d08df00b52bfa3fde3c98cb37a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"4150d17198e69346f7abaa9c2b5777739432b3a44ecfbb73cfbfc6d2416923fa","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"56ce7fc6a93c33945c77abb69f9edd0b4431f16e13466d2cfbbe12f27477e22e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"50f426d0cdedbcac270f59d003ef7c3c71ecee74d39f40217df20bf5089cec57","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"adba6a0d78c5935ccd5b1f39b6c4ffbb9fd6bd29657da23bfc25725c0fa627f8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"b47e97e61c56bc51593e436d999548b2e1fc9b1fdf971ccb96389cdf362d8790","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"15fea774d551a32626219c873faf7239bb6b4552610342d41626de6938f9d643","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment.json","sha256":"314d8c41bf76e19a081ac230166fb56db7bcaf6303425a5800aeb2f1de93ac9d","bytes":5337}},"requestSha256":"d2338479dbab8edb11a2b6a622b6777f8a693b7db2d31a011b53c1e89f4d0c34","metricsSha256":"40a219db362c9496d6dd6f6cc800f4cf4db73975cde3a750714b608df852a441","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2479239","recipe_identity":"160c47a62c51bb917b630c7f133d21bec932323a84cc6a13998dba298a27888b","loss":3.645326117693374,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"6ab36154b6fc2aabfdce9ee5e5735180a9bda08cda97c7aea8e7a6911eb1bf65","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"38ec305d96235f678ec4ef0efdd47b510c6fd0d3e41631f10a81058b6ad023dd","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"1ee68c947c0450ddb7b5b42e2f0179aca96beaca9e9eb41eeec87edab583011a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"03ca1c3c4d7f7afc07c58db9f73520b287dda8ef7e3983bf27b0c66d69b85e54","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"cc91cb170adbf334ad3d19393f5781ed4c19ec942f945c787a8b1805c0b885c0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"19d56d042dc61562bc960bd19ff9601a8738cd90fa0638e5c1ca3777fc29acaf","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"371517893a1098b38ebc888041fc07fea42e0185d96ed733785bb756a2dd36ae","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"09fef10ae92e5bdae88f20e8d8d7cba54570e899eed264e5d07283b177212585","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"735252134886ce29c2da4bfad771db2cde480ed59d4e01e278af5701eaa4b985","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"d45c1669e78576ed74d272446bc433a712c1f51092afeb6eba081aecbcab0d0f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"1a4ba75d27894aaf816479cafbf82c529ca4de04409c67e0ab09b6c2a4bc8263","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"210138818057c5cffdc777f4495e9fe42efed322dd618788f445aecd4b261c05","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"9a62a074fcc78e3c50b32b4c720b04e61be51b9958c4232bf9d740473fa647ad","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"cd650590d76010c3d5eab19248915e29645630cf582591eec44f3bb89cf48f8b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"a7776af3b9556a32bab441eb195ad2011a798bb802018a24b4503538ed10e147","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"7d9568bd31bb4431030c6da96cbd39fc1befdb02334fd7e3584af2a8634f633c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"080d9a57ff0208892d20219eea7e1936042a9931e817da8ca6e007748dd654b5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"54751448ab6969040bfe08b7a096e0704746b8c6a8a6a247f0d2e6ef44ebe36b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"0b9745bbdcece4f56b7b9b07f8f4e3361689ff22cd784cc2d810b3754f69a7f4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"49cc6c5f7cd3ccd0a4728be406056008951943717fcf69666cb6dc3baf9c83a1","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment.json","sha256":"34e48163267c6a661d82cd5befcf3a3f5a43c39af0c633c5f9def0552a0b4cfb","bytes":5254}},"requestSha256":"2f5f2ad85156819e9a9dd96ae6329d6f11cc372b19c0a2235641f1d520a6bda0","metricsSha256":"510b185fb8137f6181d4562e303ce8fd729df8763552a73cf5e77501a78b0253","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2479240","recipe_identity":"990409b96ab3f9c2730ad7907396b27ee3b5727f8873c5a1f594e8fd7cba28f1","loss":3.638544595733517,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"a1fbab411ee6c9b12d361105a6c515c817e9594004ca3c7b4765a404625f6f28","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"d891934fda5d67ad926df864fec052eed216f32fe8edc920ca5cfb45e661a302","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"afb1dd99cc450ba669312c803fdf22059b4cdaafd991097bd7eeda625919b2bb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"700fa30345a34763fc1b943b1f9ae1496e25cb4e372219c64c26eb624826c3cd","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"00e2fd9ee60161fcda51cc575c9f96b697825c89846b57c3747428685c52adf9","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"c970a8376eabcac3726ffb4f69c6f70aae29163ab42b432bc6e8bb6c14dbb1b5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"ebb374fcbc19bbcd2c1bbfcb6fc64b65f38873307ea9b9edc9bce8ab90579546","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"dd5157905cb85eeb8b2e0b41e96d4735ea5bcdebdf3d6b663846ff747752eac7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"c34a0a873e6f0f3794297158c0477243644d61ebe455244bbf56fdffaa33ea42","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"da916c21b4c4da219b76fde85cae3f52cbd5e53f9dbbdd08a0b8767af0bd3a8f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"65d8a7188896910ee0c9d18e3fac43dbc447e96b2f183cbea190046212c621c0","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"1f2ea07a4d174051bbad57630d799d7b53220d9965082584ebb80df1feb5f7ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"c8368c96fa9d89da2af9ca18418aa7f92a5ff3a0855333f9c5de2534b5c88b12","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"4445727c3a944a1d923229a700b5e325566720c56832344b09c09fde11f4be2e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"e7b8ee5c03145ccf990ec2e90b1bb74811d9f5e976cb35ac5dd0ac50e54fb42f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"a03cf4b7743b6e40f30e7a4b5ae8def181bbfb4c59dd76b09a4fddb8f1c70675","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"fede7d6dc7639ead9d1da21be6bef7c3f73f5f70098378bdb2522e2ff1695d03","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"78eee163cac91c5c9538abc28118b2e9a23958b4a1e17516cdce437c8289938a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"164004d5c62df2cc92240b50d08af9a70265f9aac864cf27a74f8eab4bd5a594","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"fc45ddba48c621fdcb770159f95d4b3d4b00c68ca547c48d3a8ece4217381f3f","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment.json","sha256":"45798094ce6412c96ddeef7dc137b0c89ebcca434370925b9b270e4a13fb1267","bytes":5340}},"requestSha256":"bdf19ffae88e696039718cc69db3b11848cfaeb5a16419e415cad204fa65672b","metricsSha256":"b85396220753cb20756a85f4a544e0d155463d498c042050fda13f5336a1d066","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2479241","recipe_identity":"28f8c4d8ba90b0b28e2122924b1062e1e676d7ebef74618b1a1219c4be59ec01","loss":3.6699918494163555,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"fd04c6253e2850b091ce503dbdf552d82b1add872a9f4fc00952dd5db799b0b0","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"39a1d5f40df895e2ded6059539d16d7f5576723fb8329e09789a22037432d2ac","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"1cb8d10bbfc6b86e6e86950ff2fae9fd5c3bdfb744011b6a6ec69ccf1307de0d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"162bda1658827afe06d8faf113744f48cfed8e29db76b25673b969c0edc25d9c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"a311cf60bc5e5a2e131c9b0896ba66da309889bef1e622e0f158d783cea69c6e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"0c787db50beb3eb7be084dc955159b4b32201e95322eaa6b70fb2b49757d59d3","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"9784192e2d97d4afb3e8cdd45d12717b57739aa1505dc32025b82eb2a6c2f951","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"0330709d9f7422da744e043ea1a0b68644f1173b63fb21128e6e0f661c40b7d1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"bb07821c9c986dbe04daa4268066490de5035301d513c52a6cf8cfc0c3217357","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"d40ef9e3de9d426400120a4c05a87db600cc3cb851b209fe3cc4c596b21c6ba1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"c5b291749bc0e4b87f3a8e819da082462380da74ff75b9ceb7974a75d29c61ba","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"29d7655d307f3a6ddbb3cb096b65ca8b3094a8223de0cc88b1b6b615554e9a7f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"88a7db7d11a2058a4a52b55cb4b1ad32e4361017b0bd3cb02cf4513f508823e2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"ac2230671b114c39afe4265b0ad3f81194e4f4caa6274037ffab5070006dcaed","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"731c97bd0eeff8e0c8cd4d6f30b6bba5e0d07cb95eed314cad00ad3f905468a1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"e46415297ad936f7280b44f16bf285f029a526b519351abbea8200832a08ac24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"64aa4012deefc21cc09d9f5063d2e856f3111fbd479069576eb73cde652e47ba","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"b312ffe811cd6ebe68f04f59e58a28751384f15301d6e35ad8a8bf0464efec0a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"43e7aa7290261e8fca47a98641344a1d25174ffd4a249880faa45cf5c87d38d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"88664853f8866ffa3b8da328f5dbb7bc2429aafac2904a111c70b8d33312edd9","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment.json","sha256":"b5c9ebb464f543676e9c889294e62e811a6b45778952b1f5f6d9cc2fc50c6217","bytes":5255}},"requestSha256":"49e23f00d8f5d1d4ae62df6d8b601a0c66aea348c2ffea8b3c6b98a23469e742","metricsSha256":"e0e57b2e9510b3f37d6911989d151689e20e70be95f59e1da57680405324c0b5","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2479242","recipe_identity":"4b50df550a7f6bac0c9d74c43fee721ad30ed0ad14c9cb3a12e6b5b68d017960","loss":3.660008063284621,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"d5fd54d57e0591f20bf2618e6e0ab6b9af24dbee5e4e6e438a94b03f251ac5b4","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"8395a97f9f81441d2b5e8bd8f57eebb519ffc30f07f6fb0198f26e56f2e30c03","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"c3d17389133a32ae56c4cce32c8e7303e79e1457dd5455c4aab969685eadd653","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"efebba2b516c342c54d9da0d81b1a57203263d1aaa62db8c68d481f440b2ca93","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"923cb1499de8934e9549fbe669599e232640462a290ceada3a54ef15bb53fdb6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"f47fc545a89133b03072f9a987a38ac35cd220a3ca151e94ddb39767cc1af46d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"b8fd9b71cf3c6cddd59b41496a8f3edeb16d88d6a2e791a01bbc0e4cd6c167da","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"499acd6ef382dea2e0a0a3c620579a97454f5069136a46664dff9f8f8b297de5","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"267229b53d6820db1b4752efdb89ec36b0c95fa9ccd1c94d004cd41c40537331","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"e8519ff9be92cde45715221e0b8892491d47b68f904068e55712b1e62d3cf494","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"95b52bbe0ab9b4b62f3f0d995be72ee038279221e1bc382f437f8ab369ad5c5e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"a7c59986a4aeeca12809303ecc3c6b0ffd782a6af80e885071092d57c87d4a7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"5f71d8ad5dcc7be6b79a288f6140fb57311951e89992dafef45555155f78a6c2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"58335bd8d1d70a80382e4d1d5f3411247fbf90f9287dee6218fe7cd2617b4d42","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"603ff344d4ffba1b265e6cc2a61dbecae4fc177246613e910bb21012ac50e583","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"c965d4289197b6b2e32557b6940e2341a429dddbb73620377c9ccdb0f96f4c6d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"606104c09cb7dc6c6021a24a7434747aac19b7982834de8691bb0f2db0aeb2a7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"ab7e1626d059e3cbf99fbae6e7af030df011648b82da8f3a8a1db13a1c6bfbdd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"98c22bb5ab5224a5351d28295d177d2d1c3117db6f48725f4465ce09f3218b05","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"a2d01505900d168ce59b4a2a295b5f4776b9c0de4f6548664708da37d9a13fe6","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment.json","sha256":"e6779ec02c8e0110ec45ec6cbb348f52d7f6b13029b65fcadba439912a54fa0e","bytes":5340}},"requestSha256":"d1bbeefc43c94fb4049fe74ad72b1f88e06e2e78623d10207569a299bd8b47ef","metricsSha256":"37c7ba02fd5e1234c4d4f619b58b3a4e0577091756ffec371d802b3e6531cd58","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2479243","recipe_identity":"33c4376b47d94666e35d5a454a932ef1f8d73199f984b3fa26591e2f236772a1","loss":3.6688880052118806,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"54974841fe294474e0b235ad5739ec4af5ed639b9b5e0c066c9d7b7a974c7cdc","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"95593ff064967f39ac234db1f7c9101ed5b8b48da672bd41e9d94bb4d3e65781","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"02e69cfa1186e8420e9d4c21958590fc2493291e7014961f54f21d0727b76c64","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"c099fcbdddcaefcd9c9673483023de696fdda3bdd01b0c43a7dfcc4e1ab0caed","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"bb05616ead5d99357101a53bb91670865c7509a275407c45b26562c0f6fd81af","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"c266c87a1c477c030d501780288739a081cefb1e49f6b4bb188b23d5f27127ac","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"5d497bf5aed2ec1feef1d7f09a6727bfeabb75a8c60457366062b7377c95c40a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"219b7b889383fce88a6cf499862a0afe2c78662652476371ccdb390de0644bb2","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"4cc3c0ab746dafcb02da091fa9df1a485995a24b19b6ecc6a7322a102abc40a3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"893a2d1b19b5f758381e5e95a57d2a70b73c43e2327aeac9e7805240f9dbcc28","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"d2b93e362c0eeb3b437b263b0aaf9a00029b8453146a43d8ee5b6f1a067efc0f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"25e3d12d0b401f8cf8edcf204d080561618002c2f09d021782f757aced566493","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"3a4f19c6d473499048c4019e20d160a8407f01f5b835abcc539b974f58109e68","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"511cd78121804a590ea0fd3a72830f3dd47e2248ca9d42e0077f819245c57556","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"da6a9388830a919ca4e6dc9b852d56d2b298cd63feff1937caec8e22c0c6f1c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"e410d784c0f46aa63c930a1beb16b98a3fc9d11c4df5959eee1e8315c59b0602","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"99f20861bf9e4b86345d482c2a0b823d891ac599dd6dd9c34322afd72190b13f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"167b79183fef47c0c760a811b93876514263da6532dbd0de3ae63f9306ae8c12","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"cdceb320e24ae7efbbb7d1b2dd36776980f2aa3ad6798f6e1f64e647133df004","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"9c0ef3880d9999ae89610f437aa48464e977c9d75fba25c9dd22cac7620b2a41","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment.json","sha256":"54b8096c79bdd61bdbbe007380e9bff66152804a55c0cdc4a1454f12063e9a32","bytes":5310}},"requestSha256":"0788e79ebac002ab532b16d912984f1d49a0647c9825e1e916650e9b79da5f77","metricsSha256":"debf465cb3392fd1bb7c9b41c3ee9f7b37761b832c5b693bda963b0d80427b2a","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.01-s42","horizon":20,"attempt":1,"job_id":"2479244","recipe_identity":"1f54d0c1a67cf6262d630d5e2f81f4a69ad23dfd221db25616b96a6dc35616da","loss":3.6797678497909576,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/result.json","sha256":"20f22f240d77bd7ad390174ba2d17a572be454a0cf46bc356438146b4c21859b","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/final.pt","sha256":"3bd70c5c96277cf7da3717abf59a3fba23adf2b9352aed95810a6ace350ee9c6","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/final.safetensors","sha256":"4475ba5fb3a0311f8b6256fb9afb8c8fe6d569d5703de7304be2f57f67c2ac0b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-000/final.safetensors","sha256":"04b424a8e07ecf25728eb913bd8e893cc77983eb7e7f315404d6bdb98c527830","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-001/final.safetensors","sha256":"1bbb78adf2fe83f5fd5db3a693a1d471fd929e92313ac94b5e3df84afa7592a5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-002/final.safetensors","sha256":"2ff2db8905b44576e094732a76472ebc8ecf2572347e99313f24ab589fd95da6","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-003/final.safetensors","sha256":"047dcd7a3d05da29787db3c243369134106cbeb54c4107bf113090e1c702dbd9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-004/final.safetensors","sha256":"4d20ad60a9a14ddfa45d13e49cbf044706f38eee9c57619f2ab1d8b95eda638f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-005/final.safetensors","sha256":"9beabe4745fb2e03da1a38de8bac0fef2f055cea7478e99a63050ef1a7060f33","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-006/final.safetensors","sha256":"2816df0d00177b6920e43c6c59bb3735975d55cb423772e1b8b4cd128316ee14","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-007/final.safetensors","sha256":"37705b53ee2683053a6e7d9cd514bd55feb7c4acac875b8df28fc8d7353e865e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/epoch-035.pt","sha256":"9d90f9c8f7bd4f60ef019f84b8330e977a48a80623a4a8a6b656cf9b3632d603","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","sha256":"eb0273f9c4f63e115121fcdb24ea93423be0030b0e279e9ccdfc7197c04601bb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","sha256":"925c7c1c52c88ef67db98570bd3b30ab3b10850f4dfce6eba083c97b8a991057","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","sha256":"6d8dacf4dec2de861f0a0ca116e7dabb1a0fd59dc326442c575f855b7a423781","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","sha256":"aa8d31da77300c0fe213178549c98fe2aeabaef49f2a7fac23ab5785e932ef67","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","sha256":"7506a65a7ba33ce3afde4471f3ade8e97268e7a18794df4771d130f7241524be","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","sha256":"dccf8fcaaf04adc4729614911641f8e67a19d7ea156771ed0425b582ab064089","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","sha256":"487c162beccb8c261d525fe86dd9b6c68366c34e2acddf154a1f47c847661671","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","sha256":"35808951bebe9a5d212601dfcbfa4e86155ec8edfb15f4f40593adc2a02b6d77","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s42/attempt-001/environment.json","sha256":"14376a21fd86c00f8d5c50601528a4f0e6f781995fdefb7fc7b11f7ad8db2f10","bytes":5339}},"requestSha256":"fc1cafaa9459e18d874a87f3c7eff41bda8dc508bedf20366849749894478a71","metricsSha256":"b33b69001c54da4754f7cff2d486e12d953c3c48d2a56ee7b980cd110741491f","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.01-s43","horizon":20,"attempt":1,"job_id":"2479245","recipe_identity":"d47fe12540978624ce056b401488c9bdf980676b7c41f183415ad56c851d7e12","loss":3.6754679553898604,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/result.json","sha256":"7d5e977daeca0464a59ce105bb57f22b7c2970fdee2a98eeeb8649b6e72a000f","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/final.pt","sha256":"d7518944f5fe272033ef536095be71579596d48e3324785117dfc12ad24c9dcf","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/final.safetensors","sha256":"4a5b1db119efe8fec96f74544ea9c22d382f0589093290776b95456c096e4920","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-000/final.safetensors","sha256":"7ffeb742b53cb980851877da861801223d74bec0873ea7f8bbcbbc0a57549b69","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-001/final.safetensors","sha256":"7708da97e6b12c0fdf04915ef5ee17a8e15cb7274b14937b60f5d2487189abbf","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-002/final.safetensors","sha256":"758f3de75cfaad59bcccdba25ca2cd8abc382fc0e094ab8fd465ecc9e305f827","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-003/final.safetensors","sha256":"fb0cb4c0556f5ac30e6e346aaa7f4916f44ddae9fa3b4a9003386c1c5473a496","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-004/final.safetensors","sha256":"3f682912e9b1cda0bd4a184506ff8015b1b47ad3b03cb68d4de18311f84b57f8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-005/final.safetensors","sha256":"ce137c2b102b5c2fc69314cf9355be4287852f3e2156c00dfcdaed51eb795cb2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-006/final.safetensors","sha256":"0009f3ad74b69fa055cd6cbe1b3ae6b06399a478a1216b8d6ae48f258718cff5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-007/final.safetensors","sha256":"c5ab73e9585c870f3d91f297478f89fbc911c325067f7fe8e02ca4d451280f79","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/epoch-035.pt","sha256":"eabceed0e79842bd1e38de4962dc2017278e68acf3ccc66d98b24754c964a5f5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","sha256":"ab41c83c2c99f780b279f5ec1de134446d2199db52a286a7aa80a1a29f5fa285","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","sha256":"704d93ea388db3694139b01816ee39a4119fabe9559ebe061985e1a6003ab2a7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","sha256":"9e54bd729ceda189ed8e638c12ece3226100606718aeb72c340617796b86c64e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","sha256":"4435b9e7795187fc5121f180890a11bf12ee96494260eca617c3ad329c4f237c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","sha256":"b88c0673b38c533dda7ab1072bf662a956da6c64e044b917aaae12d63b043513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","sha256":"6ce0a531aaf6fd0f44e5fcbcd502003f00b394d993dd56e0f4aeb9151875ba14","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","sha256":"71a74705003e4f15cabd1ae1caaf3b2a160a0ca253a4ed26115755d204cde247","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","sha256":"69317e02511a722194b766c9b840638bbc75d954df9cc9b953c8df0d620ebcd6","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s43/attempt-001/environment.json","sha256":"7a170aad605dfe18a7a89fc56279891a08b82a5e8dae05a5e2ec60c2ed6e2004","bytes":5339}},"requestSha256":"a94a3ce639642c961e045711fc4dd8ccb92f643fe77b64aa951cc83cab4ee26d","metricsSha256":"a471f98a0e11fbf599087c09306dd8068b11862dbe4dba6d0a22db873082426a","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.01-s44","horizon":20,"attempt":1,"job_id":"2479246","recipe_identity":"ce18197bcfd898f4ac8d438f4e2f530608410ff31df6354edc895df1a2ad26d3","loss":3.6844085278999286,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/result.json","sha256":"b019c0118e5277996f664ea6ef79a8b6eb8959b53e5ffc7b649162d19ac00e87","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/final.pt","sha256":"a1ef30a5e923ceb323504db1afc512086fbccf3d67804dff4fb1d143c12dcf31","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/final.safetensors","sha256":"da79eaa873457a6221b19ba0298be1e5f6b664b5a9a281ca4245ea2f6b4749ab","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-000/final.safetensors","sha256":"91fed30b2fd29ac32b11a42a7279758f1e9e4bd8c528e53486cbdff2345f924f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-001/final.safetensors","sha256":"1140e7fb1e5192e5ebb69261d052805bdda4c8f36c5d50bfe58b9e270a2e7c63","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-002/final.safetensors","sha256":"e10bded67ef3e7bc0b56fc04a7fae028e212dafe520394b12449d26c020d1932","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-003/final.safetensors","sha256":"0127ef065381f511f18a083c5e1011da152c219b3d5d30cd1db13ee195e62c26","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-004/final.safetensors","sha256":"2ae21997a028c978294a9dd054e4e5a994cba82ef53556d93e0e49368e366796","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-005/final.safetensors","sha256":"7616e9b05073b5fa067f276b9294e3ef7cfe222e68eede6bfd643ff8786af623","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-006/final.safetensors","sha256":"ca6cbac02405ecc86aa4e208d11dd6d75aff6a1095aa9f1699a34906ba415a86","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-007/final.safetensors","sha256":"d40b3cd8ea30037e0b0c2a521b0018a9312e7586209f2246c53a6732447e3970","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/epoch-035.pt","sha256":"f9940bcda62a8e6eef5c7dc16590d7e55bfcf67a439ffc1c960d930e5d2beb73","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","sha256":"ad4ee3ee9fef0aeb5909c391d94a02fc19a004f77cc1b10e2e146f711de2a3f0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","sha256":"f5e614cacd9c0bef344b8175833a523cd66292bfa8521727d939592b6d6eb813","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","sha256":"84746b686d027b48ab225c10643c6680057e7a1b2bc434e9d68af2886a23d793","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","sha256":"a166b743295ab4c17e7955878ab49b6a366eadd321853d116accfe045fcbc6ed","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","sha256":"54dbcf393004bd9f83f9b520d9dd13f969367b394e75c10a9036fe58907519d6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","sha256":"17d5ddebfd661e0dca26c73ecabc173252e6733ad1297453e9d3a785b4651af1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","sha256":"a0913ad7a6122918cb78c9fb5c4b81b99d0f01fa066f157fa2a389ad4aa63e74","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","sha256":"6ca26fcb24a31be9c34b8b62ef8a882ec745308a1e9d09eb5cfe9877a4669307","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.01-s44/attempt-001/environment.json","sha256":"f94b46341362294968ab1fd0d2a9244bd93a41e1673af0b12629f19454d999a0","bytes":5341}},"requestSha256":"1c9c454bd9171ffc5b8ea306513b223d249c3839d5c16792b65f7311d60ac2fc","metricsSha256":"8b66d0ec53b9d273606b91db19cea075c0edea5327ec92c19c7f7d8f312c994a","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2479247","recipe_identity":"3aa8ea4d09604cd103b38cd09aa09be1a21fdf6651c3c0033e06aa19e711caf5","loss":3.6750836838361454,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/result.json","sha256":"4fd960bc8ab3a974ecf598bf799a9ef0710f86159e07b37382ada06067729a23","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/final.pt","sha256":"81e08bd77a1b4fea738a97b887f1199f241ac83ea3b50c85a0e7ef9611b55567","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/final.safetensors","sha256":"cffda1d7f8b9829557be210a59d463207b3ae5c8b791864ecf9a7e3ab3ebd780","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-000/final.safetensors","sha256":"402b8b7ca7d7cb0ec87b29055b64c945002cc97bae08c3088ed0418472fcb453","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-001/final.safetensors","sha256":"3b1cc3ee8e5399d67a3dfdb3e8ebdf6c41b71dee58fbe4289293e9468a62c26f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-002/final.safetensors","sha256":"93f7454936a1c59b28fe509f989a9e6cd467a2a9bf5a3ce665573f39c63688dc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-003/final.safetensors","sha256":"a7ebde3bcdfa0bceebd096be6ac00ba85cd4bfaa8f1cd4ae43a23f5ffcc08108","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-004/final.safetensors","sha256":"24ebaded4e84a1244c2bf160314d2783b9c9763e9cc9ae63799edce25471e943","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-005/final.safetensors","sha256":"7c39070408639b909029712b189678f17eaf600205859e0b2745366e51919e84","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-006/final.safetensors","sha256":"3ba914a60e22e1776402eaf2208e16c01c9909fccd7741b2ca96d41c6e0e60bd","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-007/final.safetensors","sha256":"ff88d4df35dbfbf7aa09c469ea0c943a988955b6bb30e39dee83482a75724256","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"74f8074f925ab8d7983f087880c074ef77ef496c93ab3a8f4d1dd4a312c083ac","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","sha256":"f65477799795087b656724b06c105ca277c2c1f4e8cb854c405564df4383a961","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","sha256":"f7418b95f9debe9bfcc46fe6c10fd5a190bc5a13b18a2512e547f36bd0ebab43","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","sha256":"2c7a99f4afcbdfed8afada26691351ed203c1f987c5888b054ee53a7459f3e51","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","sha256":"c66ce46fdf3e17dccd530c82e5152bfbf23025c5acc6f7ebb34529a609d3ff3b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","sha256":"8497f548f68d2f4f973f332e91e8d87817a46e1f19c3b1216cc83e11855c86dc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","sha256":"18f8c1ce3238ca8eec60a77652db0bddbb19ff65612ed1c97c62a7a7f5a22fc5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","sha256":"55d1636073078d8378c66915fb6545de6ac9eb3565a554bffeddbf420e40cd08","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","sha256":"77fc08fa155692e30743e8331e5305b2ef1ec22079c06219af673dad3a6201a7","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s42/attempt-001/environment.json","sha256":"ff31a649be1e01fcf4cb58df1eec960b4ad2d4894560142ca512b31b0727a0f9","bytes":5341}},"requestSha256":"235950fef5db097baa7b31805b5fbefca50f4ca6e2e1982958bc7041d1d6adec","metricsSha256":"c6273510d0340a02e826aa24a8ba625aedee6ea0727ac6880cb65f42fa96e0fe","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2479248","recipe_identity":"74014c085d3da72e76ed7b9caa9625f8af740fb3cfb81a21fdcbc3435bd32d3c","loss":3.674464403217774,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/result.json","sha256":"366d68af1cfc31d105c0f7e096f8444a19cb616f7015e2ea30a5241b02ce8247","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/final.pt","sha256":"d01de89b14f005f630ea6e8c6a3729626031735a72240a800e7717a98fa01456","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/final.safetensors","sha256":"8a4df05631f486824e98cdfa6b94698e90d737cd599ffb2d88f020218ce22279","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-000/final.safetensors","sha256":"6c4d1f7f366ea67785d010a7de242f4c45268653a9fa2f1da4b2a711e272f43d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-001/final.safetensors","sha256":"511f9a1fe9d1ca39a8a8d269dbecff3288f83419878ff42dc1e817c41c7bcf46","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-002/final.safetensors","sha256":"94f1bbe576f884da0f3660302e61886b48e4f5c4dface6e1f8f29dab7b6e95d9","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-003/final.safetensors","sha256":"198f1102482addaca90a10e57ad2d7995c167bbdd2d576e4468d7a16574b03fd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-004/final.safetensors","sha256":"8c6d92b7db01ec8ea39223f9d1c593018a4fd40bbafe1bdca69b7b3fa7d6a1ff","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-005/final.safetensors","sha256":"b961f5c316e7f6f2c361c86648967bbfb7ef82d892f5409831c4e45f5c48ea52","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-006/final.safetensors","sha256":"fb82209f89d451e731f7cca085e9ff8c229fa700b5b3d546545963be40b3875b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-007/final.safetensors","sha256":"89f0a11da4e53b37c7ffc77abfc4c6e048b88631b933f0676b69507b232557de","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"a3124a015784809522c5af2b574265e2a27a50627be58aecc800187407f9422c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","sha256":"b870465f45ca278ecabc36709a1cfba3aa29bf592334f36db4744ec5d2de5d08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","sha256":"f846dc36c556c86b3b0986762d076ffb6a08b0192b5ecdae0429b287bb14f6a8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","sha256":"9899c2131ea1b3dbe96e3fa1e6ef84cb90cf71743255a5caee479367ced2750a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","sha256":"3154c66cd38fa34b4afd6cd9b99b1c270d116570d0a59543b607ee73849b178d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","sha256":"b129bb1b0e3bb929b1e30e077903022731de39b6b8c3e5b555b4e60be9381baf","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","sha256":"e15d94cc8ecff6606464d00697dc7a72210dbbd959709266055ded57ff02d3f1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","sha256":"b9c8dfcedaed6f86c61d19a09cd683c870f53ded7da57e5cc84bf78c78332aa8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","sha256":"6f74887b985c110cfc0c4cb73bb006315c137f637465659f8bf0c154cf7f1ca0","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s43/attempt-001/environment.json","sha256":"698de074ffa1eae96bcf2a334d8b3b86929ee25baa9cacdea2f84ec416e43192","bytes":5255}},"requestSha256":"68c9c123ca69e4eae72aaab1f45ec818afdf3b06735abe79ca2ec573edf38ee0","metricsSha256":"a306c039c03ffbd277617eb69616bff4b8ba48f33ef1f364a0bc74f845c8ec80","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2479249","recipe_identity":"80ac34b8fb9bfc9aa463ca2c892ccf42517921bcafac7c36530cd9d2736b3a86","loss":3.6784932157365837,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/result.json","sha256":"3565e3880967cdf26fd41132ce09fb72e2a4c84d335957289d0d2da6aa215225","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/final.pt","sha256":"f4dc52f8f0966f3b8bcfd1a5e2c876390310ea830b8136f83442fa1d84dca340","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/final.safetensors","sha256":"0c46b3f482143620431e1d3f6ce81df9a305e17b927a34efd63a87e1ce95caf7","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-000/final.safetensors","sha256":"1cc608a2ee772eaed74617fc695cc04f852d166605391b3584ec821564a1b152","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-001/final.safetensors","sha256":"a184326e04938864db15b8866cb2acae821e2c3172d1b9c2e4f2a776cab8caf0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-002/final.safetensors","sha256":"e434cba9e0bd3157e08f8a1ef0f82daec1bf9c9411600bf6690a9fe0fb227ab8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-003/final.safetensors","sha256":"71b652ecd2428f6480810b68da7591ef11fa43a0c51c9b5d94925a51d9576a98","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-004/final.safetensors","sha256":"c1f32cdc68f9181070c5f2b3563ca3d321bdaf64211809d6d2d379040ef69732","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-005/final.safetensors","sha256":"2cd9ed17d3ebcea264ebc8ad7d34f6ce1367401a685cbea847d449388de028bc","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-006/final.safetensors","sha256":"fc86ea0d8def43ec432ec3a33e4a320612a9c1556e3a914748a3b336a7c88043","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-007/final.safetensors","sha256":"a1484794e7a27c8068e372b30b262ddd0531e0671026c9b8d4f986070c0466c7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"1408f02b293271db9d000a3904f62bafe8ef75b672583ec39ead942577735dd7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","sha256":"f5f50fe502b764004688a48fafa848046cad2662a01320a9e3c3308f94354a16","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","sha256":"2f436575c03f563e4af50c076fbd28671305d22944dc3a8ceafe1a5c479ad7ec","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","sha256":"0fa55f74b3336f9f3824062d26a9102570cb0fd7e47bfd426c055a84dd99871e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","sha256":"27d0e40abf5087007f7a98722bd8c1253bc98db52fbcdf977ed89455935586c1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","sha256":"da027be26f7156ed363888e81391fe7240843ec8493a22c142d4655f4ea522a3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","sha256":"1934ec2b6df9cef8e8507fab7e8369e9f8a3122e0606538b95a28d63f14cf528","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","sha256":"51311d8148ee0c13a1a6e2f62b08793e09240d5238f71aa54675a66f6d7d2c16","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","sha256":"981c343bbede0445af8d5a87aa300746c1382a2f61cdb94e7349545394ac5636","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.008-s44/attempt-001/environment.json","sha256":"8fd4a65c796773d6c74ad9fcc6f523f7f6e2e8a8569737acc076b123baa36775","bytes":5312}},"requestSha256":"cd9785be2c70aa23b2b42eb631124edd3ab3fc9d14ef44b7c25e58284af7ca56","metricsSha256":"31818e9cde892b224fda5654346a4fd55c3630ecfd634b2e44a68a03db22507d","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2479250","recipe_identity":"0d7f5f55f924dafac11872c31147447743b41ef634edf94c7d3e5388c3736b69","loss":3.674766800128119,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"d95c80c3abe3f2f143a26dffc3cca2e3a076db2385e84c5c116ae91d4b175c48","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"f451ee4c9fb413e3396672c234a478298a3e53a015c1692d681930a1e4062dbc","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"15ce45590eb5af89980606a2a7232eebf07eb1714d5820c24f38d00dec80eedb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"e42b0e11f448a05f8200f5c79741f936865b1e2723af119fa3be5adce05e0c76","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"bb16eca398a53eff6fee48e02f42a4391867809310552cfec05a3624578967a9","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"9c417c1a87c06fea981631502b2c3eb5e07c3100e31f510f6f55148acafc8321","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"61080e12209dad99491a42d0432eae7b4e38c10717c9b9aefa2e5c2ec4f305b8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"cc765d46a2972afeaf5885b2fbe337309c5c163ea56673542bfc8fca1d407019","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"541fb9a607b6e443bbeac2a498c1d6b0990bb467b3b656d33634d5fb22167c80","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"0001721afad44838d62516b598e906e0309b66c53276f3930f85f546c39c7242","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"cc938d96c0d528d2ab8ed505b8eebc318a627b6df81d710979c39eba7cbc7fe4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"3fbabef8dc649b493342cda1a7b4f388163f0b366d5a850b98d209838f4614b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"7f6ca4296eef8933d76a1ba537efa0043d643c3ae92c67514a44ad6f8469ad18","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"2a016946f99db3c341a3e9d6ed8554aa53f6e5d98dfe982bd06ce76ac6f67e33","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"d7f44daa59efc8df090b6de09bd342b1a5b9a77074bc1922e78eeb834a3c952f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"bec891ab29abd0444e7ae41660c1316b85b8701f7424b462b6ca2145c565f7fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"34890751a3dc4a3775555218237335ebe119d6516a5a8b7e2f8297e15b1477b6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"ced2a25d1c7807d8450c244d24ea253d1e92ea1bccbf6207bd0117f7f10e250f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"92ed6711fab9367aaa316af8da1d25a794e9c42952e6d4e37ed95db9436e0a4f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"0e7efddc4505a3b28f7a052be1d39b0005a7f5cfe5382a756123cf1e10c48ba7","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/environment.json","sha256":"9d20b09f774b265c9cfc8554f2427fc6c0e526a0b451d6fef6a597a961809b0d","bytes":5341}},"requestSha256":"4da4d3f39fff4a9c93c906d14bd5429d6edb46cc9cdf3ae09a772d1f5de7868c","metricsSha256":"512a1c2c09c1813778fed4a0fc68103ed1ad3b0ac5ad5b5a6514c5fe292d68c4","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2479251","recipe_identity":"9dc0b6e9e3012ec3ed0a1ebba8a81e1e83e9f9eed143f7b7bab89edcdcf2bec4","loss":3.6807401945053195,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"5f35bbca87abebb4a45f9ee305d9dc58e50ed04a00104b565c8aaffa06a25f1f","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"723e0b64179d374eb4f5df91db063eb415a0d21cbead8107e8bd4d19944754c1","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"26b9b610e31497b362e177835afd6b1849604f903689bfaa96c1f936288a4e6f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"03397e89056273b5adc844dee94cf95ef63c90c38926aae198a361237f8727bf","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"30539d687598e14ec4e4b35526d95856421d0dfa7e4d5ae6a592c0e7a528191f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"963e022804243986b2a89d94bec717f0d2b2b73485ad51f0b043bd6c2ef1ff10","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"393a1647d5c4a3b56c063fa36f65461f1fed06f2258bb8651c2cb0aee6098b69","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"07f5195050e6fa877205b6677bfecf288ad347a68aea30156911338a5a8567a0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"0ed6401a426d732565762c8b7e6d707c1f57681342def919747c692781082fe0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"7150012eb12abf039fe276b8ae33e776b14dd07ec94de4946090086b3ea85b1c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"33ee1a4082af70e782e4d2c666cf26b176fb688d7e7d5d328930622b3fcdce82","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"374ddf274c59278d34a2899b673e0d505a8a73d3cf2f0699ccf2bf454e972a40","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"38bd924034d9b15e4430324e363a4d79de96a28ef7570adb3dabd63b12a6006b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"22d78a9621b7548b78e029a4788c125ce8b3ec658940e8e441d931d070268e7b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"581eb5d6001a694609499c30f4612daca7a7a14e516c7577212fc7605370238b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"2a4ce405b7e047a944f18f14794fcb4bca9d89ceedfd8d414f2f4efd0d8b92f6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"fcc3862d05608ccc252c729865d43fae28b6b7e01debf14e4f17e4e3dae8b785","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"9a1a2d2ca4ad3db580c6f59e7a0f730654b7ec602808aef66ddab2cd6e12b4ba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"65981acdb8f10395994ac5f91ae4d5a045ce46c9f05b359df9e4221e3112ed73","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"fe7acc170e5fc16620f9113281ef1cffb2799c6f0f88a78f0d81b917bfdca880","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/environment.json","sha256":"9392e2a3c39495ad7537b3b42aa2c040542025524c251612a2989a04e6e68fb7","bytes":5255}},"requestSha256":"7c71ad3b5efe0caaa6999a6db4a8c80b34faee1051f705ca79d1707f23e590ff","metricsSha256":"663718265aa713903ba9c9fb68271727a92e52e3ce6774960443fa913a693137","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2479252","recipe_identity":"0bf305808a5cb5b2424ac98b9e8a74f0e921ff024f4e1c706cc87021f7d48fcc","loss":3.684643268230684,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"34cf03422093c149867d6dc76b423467f50f9ca098ea0aa8e8055f1b2b1eff98","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"167acba08ad2eabc6c216abfd413ca32bf0734012935471369b75aaf9ad2c6d5","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"a7d4ffc1db00a4f2ffb0acfef133aeb690a6350022e5d1bf61de8fbdb9320de9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"68e891985fcfcf85923f9ee66a8547b301ad76ba3676ae1f4aa11d04137ef05f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"63ab113d7cda583106fd21d72e5f4188c9b4ca2d219c2173ce72cf7b9f25c9fd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"16e0a266273095f2ecfdc63c899ddfd4df224bb655175bee97f427e371453ffc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"04695cbd4f7f0a9c0932508b8113ebd1d24ec137bb7e2f420d2af12e0e912177","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"64e40a12b279abd0d0a2fe486f0602d2b14797c75d32084aefa68dc1d627554d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"5005865825453b93de38baab4d28bd304c17bbe736e8b9c06e739f620483c108","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"fe20919c7e8a77027d19adebdc4733778f8e8fe025bd3b775715c9a07431cbdc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"61ee3e20dc9b665c7942e43c4c669253b8119f75a982c538bd7216749da2f056","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"e7bcc78f31a56a4922c35f35617466b50a97a2d1627213242db67d39e37b58ac","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"723f672f73349c87721985a0f6e61c803a2db0488f8b54fe91fa08ea1d5e27fd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"34ac52f0ed7dc6417bca87894a8a648c7f42ed11fc0c081c07eeffbc8703c69d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"f93b05d0d1f770266c4514e3ceb7b8cc92f25e842092086eac1303c4b3e4d8a4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"c46ffbf80a19fc7c8a13e1fbb54945bef4747c4cd818dd3a2edc450fdfc90389","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"a586addd30c26c09c35699a20c4c908fff7b23ad7554102747f0a189b5e88d2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"93a8a3f1534c2d8e9843b16c5cd16aa838291db96c4f0dd720bdb8acb0221181","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"08972b96801d60a4549645d607062a9e298b1f771fe5b12d02658d65575fb773","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"54a6b85a53e72c7d3c17782bafb250e76ea341800dd3827051f167fb1c91d435","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/environment.json","sha256":"e07ae1cdec1634066128b8f8728bfb5041662c00f0bf27009d48f99a6d4e459a","bytes":5312}},"requestSha256":"5db42add2bd537e88bbb2942072a37b7206074d2c936c17d3f152e449b69b427","metricsSha256":"2de5c162f94603f1439f5dcb453b6d81d36d51b36390803ac356e01c359b61f7","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2479253","recipe_identity":"9b78a60c8436fd0ef5fbd3106d51ba6701583f3309db5b935cee444396966e95","loss":3.6706539413275823,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"8ddc1909f1413c6b86a9bbb16435cec6eba42fa44575c29f87c3a29b5e98a6ba","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"39b444fc0606b4a75d9d4078f62a23e154cb4855c19997fcb6dd9fce2987cefa","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"8429600a7a1abafef1255b6c6c80a817dfbef7fec0fffaf7565a2658dc625c17","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"7b0a93387ee0217b7eaafaca0ae299b9a8e532b9444fd58094bf08e2620ae294","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"fccc87956fd4b0d99a820459e6af63bebe5a16c079575c61268a872ea282cdf6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"303931f22b89a9beee5124e3561bbe0d153d186b1e2e9311009ffa32f2c05703","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"cdda963170a4d53b8cc427b375a41e9511a621fd338fa4e8f59eafcb2d325bb3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"35ab2c7f32133ec32d89e8e182c9d4b6c24bb6734ba68446f62d8c97d7e8a27a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"e6fc9d63ddad6461ef3b03fe54959cd715d263ea3f0b9bf812dc21545e3a375a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"601a8b3152a980af8a020308899cc6125fb563c599b6e49712ec0e8a6ea61e55","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"19732b3c33b58c3cd50ea54e4831c9b02652778e7e00c2d43dfba0ba434409c7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"1cbf195798e0e8fa577f41067b8a32c0fedb98f787d82339c618f59c0da79a80","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"26d059fdb835b9ea35e91e2c86c3755e280bc9ec18253790d7dacc5f699776ab","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"15d67569dfa6fff4791bc6f7054014c99f81c18fa71ae0adab897c565d23006a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"ae030d4161144f0e1dc7502a0022bde3fc82d47b68e3b347afc3a04bd5eb3734","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"2d1c07d2c3fa1deb5b654f952aef9c8b4dcbac1b6e850a1de323579b1bc80a7f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"32ad4984fa47ff78cf76cb42d8e70ba118fe94c7efffac744da7f663edc01421","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"fb484f085bc7ebda4d19ae4441c2796c10f57f32ef1b8c32fb81b05b220f21bb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"7259ed5a24f79752e12b48c1765459e89d53aee53bb8fc3adf91128d6ef9ff41","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"830fa65428fbd9bb73e4c0418948b04dc7e7ff3d45814e8fa732b58ae0854143","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/environment.json","sha256":"d0a7915bb66c525d03f855fe4e4c6d207b3c81ddbd822b17b97c9beeb600e890","bytes":5341}},"requestSha256":"1a36903b892787134e3feb9d8b7e39f2084f890ef63810e898a60520529ae08b","metricsSha256":"a88eee33eac7b25d3a747569528ed4616eb57ac8b149e8ac3daef7d5bc60f1cc","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2479254","recipe_identity":"400eb85d6d24e6ea605b857e408d6f79853cd856ae6fc343b57abf2758d03d92","loss":3.6821017310288564,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"123bccb70d4b25b0c2a6b366ed351ec197d7c91a9729e7cae2a1288472c087cc","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"8ce0839a8d5ed49a942fbea56232d85cdb00dc1b6e4c6d3054a92bc981645998","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"93fc3766d209d7cc3bc3d025095ddaaf81434f4ea781ffcd247d86abfcedf0d6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"902018cdb5aa898be687f815470f69c67717ab5e1401fad5a27a6b7d7d7bba52","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"57a56b506aa085f675a58c35e442f587547fa51863fdf772982f6fad388850da","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"747af326f33f50e2fb8068fb54ec752b81f967e025bda648adb61d4d58553c5b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"f38af51950d290672cafacdff2af5a7b8a0986ff205b83c9791188f973b8d417","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"af1e2a86c2ec728407f57b113eeb70bb005bd248e43e5fdc964812c41a567622","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"95b87e3f68af26b375c4bf222e43eaf37a60945e552cccb8658f9ea591de206b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"71875c164b62b76e3f81a75247a3b855bedbf41742fa9704e64a0c5df69eda48","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"c46c0d5c2d5eb2156c4db38a3a1ede341050226110bf8747da92f8d7e4eaabaa","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"9161635c523461286817912746cc603a10e209657abca2adff827865ee3f541c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"a07592d1abcdd0f3f7b2592a4254a71cf238afe1dee3af31e43f2ee0aba5fb16","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"f3fbc3d682cdc7f8e25b6e18b22b105ff7b1c2a018ac17e9170aeb9e4d2c7060","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"504fb40938aec512a2489d8f86e44269205f2dbf76b0eb3e3e5ee28bed4f8bf9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"3b1dc25368678ea1e1c0438e878f715e821529eb3f99748b79830fe7dbf27045","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"9dbb735de493a23181537c0534dd29b399c788a39f7dc40671bb30e61b2adf6a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"21c048c7df3c9a21760f5cf6e1b0344848dd333ff7c942c0922b333bb210881a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"052a43b32fc2e315c7dbc1c6baa5a09ecbc5c88593d96512f87cb025adcfc8e2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"31c0a78316f8ff8633134b9eb1afb456cbdd9adf64da591fe2fe47daa3fc63cc","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/environment.json","sha256":"223316096b8c4a87649e6c7b05ee0f29941584513efb4e99d15f6b2ba587dbba","bytes":5341}},"requestSha256":"d86ffa4fd99607e825cd28ccd1982c57fd418c539fcee4a963d50e4991fdd94d","metricsSha256":"f7d0ceaaf1ec61993fdac194ff860e2e4035c6bd33125c1cd799d49c703c26ed","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2479255","recipe_identity":"d55be4bcee03c6237a0713ec344b406528cc71085f5425f2ef20a9a44c0d4a10","loss":3.6828935666665137,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"06f147fa4063fedfa8efa6b357a1a3b65a3d3d25159582eeef248d94c398e1ec","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"8d4603a6a1ea1c2f443b0bc152edcaf0138995743c4a7fe68f4a3588c28a28d0","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"2157d908e5c8be8867d882438d2e14ef07e444b139d075aad9bc38b954177c84","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"29f7a5cfa05a9206a94ddd2adf41770fef755d9d261da18166100536a4e5cde8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"00171f286261bd6984809b3ffbbe8a245357ac582a767bc53a766a1c1ee5039c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"3c85304d6df297112a599c7970e84c4dbab555fbf78184cbc7eff01eca4d941a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"4425884eb3a3266ae83e2260d620c2d0de1ad792f2a92741193b3f1a62967ee3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"5194cf0bc64c8fe13c111e6bcff26e9c7287dbdfee64ccbdf2ee9de2f08b39a3","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"0d46e463154b1bc542bda0bd3c1089d767a1ff54187dd33d1b9499c1e5c3c59b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"8230dee5e401a6eea561bd1b6d4d1c556de17854b7ce852253e12db3605c4fdb","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"ceba6a0602567eb0442404495c0436cc0f321ec9c97454196c3a9bce73885dc4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"c279e159107777231183efb7ab8a87f15fef38c9738d6a6e23e665697b5672db","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"2e0966583fe669411538f53e2902ecc9ca3bfb64a9ec9543ee43073258ea36c8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"138f78b5c3705e0238a5e698905f3afd55ef0f7be24bd7367bb95d991a5bc2c7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"e271a21dda4d40aa666053e9e5b656b05e84b86afd91eb791db15c53ecbc73a9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"4b229c31e2e86436820d3a95e8b41dc42aa4d2f9ea06c5a8e669f87ece7e3d24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"36d6f3ac5c570e2f4e840fca1cdd28196d0e29247a6aaa787da79a3641fd891b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"2cee3ad05c7ae2fdaafbfb5a77f808e8ccf26f22217aa71b3f7d027a3644f077","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"6940d890174fc6d504ad99c5668153c3adf5173e704e46ec32299bcf88175e4a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"4c7637caa3d78090bc0a4f7e9e84366660f2a4a8b6c6ab9f3d4437b2c73a6cd5","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/environment.json","sha256":"79a4cc12fd2f8b5f56ab3751fbcb36191f898fc69e7f52990892f03ad061fff0","bytes":5255}},"requestSha256":"cc5914e7d1aeadd665517cba6f6ace72e21095ce3bd7b3f38f584bf82df2a04c","metricsSha256":"5d1bd6b67258e1c7333374db080475b810f843bb49ef05eba8e3ff4d534dee8f","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2479256","recipe_identity":"e6b21d3c15eee2a62124cd752243a1ef45eb7af94e6eeb53cd7c8accfe6ca955","loss":3.6842535597515753,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"6bebf9c72c5c7c8d0844f6f36048748af8d332bde1f9178b54f82fb90beba20c","bytes":881},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"770d13a2d13b39c60e4d04ff37a6246ac3049dd4130032c63635536366a25aa0","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"86f805e111a1fb245edfd3f96d5057b3dadb34d80a7160f7eda8c9cabe49bd5f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"806a46cf9143dbb5359a35ed77607d23127bb1ece8951d894d1d03972f0f5412","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"dfcc95bcd9eb5761b3d8d1d1cdc5be383aa02b2699f3b82cc32a34d6d58d8890","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"e7d5e339b5793a45ecb97bcf374118c0bf9f27c33e1ebdc3533c44492e458452","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"98166049964014ea62d4fe6f147fb80e4d8ec9b06cd5082d24dc5e5e82b876b8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"b15b2cecd56f1063aa6aead0422ec6f347e2c01bf679f7f488fb89718b45fc2a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"abaf57be57ac7a6c00b35e587f94b94082543f61e4f393d2b35587283da352c4","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"91e996c519fa15993d9c84395b8fa47d5bffde6a990563c7501725e25da54572","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"f20fc7a12408b2599039a082ef01a0923349cad78c7eeb932ea42c0731101709","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"8346ea2b483c9b4a87bc024bc4535af998bb3f83b27c91b0488f3a169260b8ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"d17d7210b5ba7a1fff66d25e46bacc28e65e61179a79cd27c46e0e1c4702009b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"39635d177c4842253aefd0d319f9a585a0a7224f974f3acbe449d24bc7b72b50","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"334b01e5d067b18a8bc188bebde53db5e07aad88056082ef24ece9a6e2188da5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"a2a14f8471a326b4e5e7b71e984a68b66b91a10e095658d2eb8ee943dd6898f9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"b8dc00aa4f7481c795f5400621adbe3537d2e580d7d98c85c8ead64ed30a065a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"b63f342f87a11f026d2ea2ecda53a55befe89ec783e825b4e7dc4825bc3088a1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"08774e2270f57dbf6fdcce909078a026c12effe1eb07e69268b1da953741a9bc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"7c3db5fa422287e79c88fa21e353093756c9b145dc5b4a31653301fa85f5e7d7","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment.json","sha256":"93d1e43bce9e938e9016fb6924ebfc2ba71b0e84a97ef3cc0025945848f5fe5d","bytes":5255}},"requestSha256":"bc8c2fa9a7376254a4be19a87043c0a6ea247f7ab55f02f614f5aab41367bcdf","metricsSha256":"8c30bee8fb7cf11e7cd39cf71fd9beb9a7af332b65e9ffd8d7e50722f31c6caa","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2479257","recipe_identity":"984cd34056777b030a527392f27423dd95d8a5c048cea43e319f4c5daf4b2b29","loss":3.6910077991873904,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"c7321f3424bf78cccf8299bd2378ccbafc33390dd2d05b70d8aeb388e25cf53d","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"ea86c2b288839cf28412358c7f74572d4923cd2c0a6aab5349aa8d403ad2006e","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"2bb90065f3caa3ea6b15988698b928ff046a992628f84a640bbef0c4226d83b9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"be223e5cec58516f2a40df337d4d3463ecbab391d2bd44f078fbb1f805441ffe","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"9dbf3b1ec82d72930bcd76d8ee266d27ffb91d3559c6a28bb2e3a467dce39982","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"a5a6b69fcc39e4e199cda973a2d6b62a64b453c511cfafe9e2ea04dc458d7e2f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"661446078d358b55e146101721bf466517f364c9415f3e96559f72e96d3b551b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"a6db84707bce76d7f8cb9eb19dfb35299e99174a723850b036a8b89086fc9ac6","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"b9dced0a2dde431217573b4c21e05e60688eafe9820bd1724dc58790cc8bc543","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"9e0e74452fd599b15513256fa60ba36819446f74e1f04f14f1a0cd36e69a1ade","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"a4aa23f508b2b5ef862393a45d5ce00ab9fcfc1a52c74f989bc9ba4fce336667","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"bd76947f89a8d1300f0c14249881f8848f778eba5f3d46e6433c02fdea0379a5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"29afe611f5aa882e281869465a5af2db3861b893841710220297642a84bd2d4d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"22989395a399d392e10e54523f83d2c272bcbdc4ecd4af73636716f4e614b1c6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"90a9687ca7a89abce275fd977523ef5c9102d79dbf0a0e1d809605d03637673e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"2e044bfad7100b9fbfe97782a269bf91b45933f0bd7d8dcab81748c2d542f350","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"6843b6e458b0eedadf4d3b012f439d44c302d4b39a5d46a37cb3fa61a9348482","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"cb8c84d6c0b39b8c5769aff8459a33bd290e2680489a0c100385e257023d5f65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"ed1e72fb4a937f711ac6fbc3fd5933fd3e599f1f6a0685433c05a1b531e0b96f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"31267a65e3f92f998cf835b5987d058369b5bf9fe781559e14b4d03ccb34a592","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment.json","sha256":"a20100c0578f5032181afb567de577a45a7ec7cfb1c5438b5ef34d5bea8bfd6f","bytes":5254}},"requestSha256":"6a38aea78c04ee90fcc6aaa60e938ee29edb31effff00e82b582d972c9cc2912","metricsSha256":"a0395411e39f4589047b2b18d21c8bfab16deb2982e6466494160608f94b5b00","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2479259","recipe_identity":"8b6e4f0bd34392e9bad8cf5928daa86ccc1295a6432d2de1d0b7afece399fa86","loss":3.695541996941523,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"bbaf85ce66fd89244ee6d8805f201ad468c72af3f98981e4c9c9bfe6fa340888","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"0f3ec4978cfea80eb52535d6bc141baf3ebef5e33c3aaaa42d6a3a63eb706bf9","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"46d4e90eb17fe0668f51fa75fcd42443fd34c6629ac6811da1048139d8ed2fa2","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"9bc531908bfc86fdd0e54e6ff86c59889f83f0b31f7125335dae706f6b4f6c2e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"3f615c556658f360785ee2b126c1a040c6add0d9499a14b9fda68bf34a855fb5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"edbb422ef63af6ecacb1edc2cb560992c80d8db2107ea852ea7452c418255aa2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"681a21b594a44e819537b8a5d976ae105d2cda4e6bbbf0025531809a469cb3b5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"f89d90767e9aa9b26578eeae3d36605148bbc55c2b6e94743956d6b237a24d0c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"7cf3fd65aa1d601b99c262fa22a5034e0b456e1a5cdd07c6513dd6355132bedb","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"c095f606ecaaea55b8347f26ea02cd4bade4edcddb034a75fde3b11bf5fbc965","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"ba7663d69d2700b1f9b7e0d858d181fd40d78f714e2c741533b0dcaa80b2297e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"93af5e0cddef2b30ba72738b7c9cb6cebd5d210c7811e3f73d5bb09acbad202d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"82244b7dacd2f402cb7efbecdb6e41e3fdcd841a0e9ba4427725cfa51c0e3b15","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"c3e9143c6f95b389ab834b6e668c58c16e1a71e45a24952949c1f24344ac6148","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"5f2af2981078774c84185847fc6f3d912dc99e05da3e71da0e9057ca6916cb0e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"44953598b12a9ff95b731a834fb869cd38ad768069db78ef76d79f0b98265300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"15f9a57eea8241fd6a9d6beb587fb9ed338245261cc2e5318efa6b08f5846ce5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"18458920691da003ba31dffa6e6c65f61af30aaea47f914ab82f14d3f31aaa2a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"ea1fe629ea59869e63938244922d622d28fbbbfdd1680649bd75f1ae1dd8100a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"4a8a86cb9d4b6b58a255ca9bd73c29e84d59d11bd92cd2bac0afb2b9dbe7f9fa","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment.json","sha256":"b3f456b776c715e9078134a136671fe11f19c5702836ee5105d8491b1a4f405c","bytes":5341}},"requestSha256":"58665db282894d668fbfb5a6c55fe19f523f24239ad05e913cc12c3c6b96502e","metricsSha256":"cd21ae661b9bc6c699f95df5ffc340aaaffb631639b5465db7b55ef26cb8e7f5","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2479260","recipe_identity":"d5b46747651389a2a7f04e3ddcd101dce69b17fb70b28f58d902eef02e9b5dbf","loss":3.71129650006366,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"c370491c887e84e81b2ab29d7107632d478599d887776aed11b35c1b6b95aa36","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"6ca615a8bd3751a6a2b96352d79e24e43f5feda12acb52ec2e275f56b4dbae24","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"ad9726ef8c62d1fd8e59339d4de6389aa2654cbf85429b78a2c644d6d62234fa","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"da3f5b2d997953b62671ecd6820ffa882e9035c68faf67964600638d676bec94","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"0c55ff0996ef49e902bff29bf09b6d538314d0ffe805688adc30a9e412b1eeac","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"6bf92058ac8ffe3678a39ea39c70a270b97e1d362a4cb9876374886d168e076a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"7489d4ca79eee8984664567571532992bb9d723e78f22aa5f00f5b094ec9cd68","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"af2ebf4bf5ae2454853cc63f76415d617f6471c23a0b0f9254e7e70c13cda6f5","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"0995f460c660a8aaaa1b7db72fbc25bc393ff234d5b84b303c9aa271cdaea545","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"c49f1d18c2f49415c7d25b54d6334ff30d0999e6f690313f403bd43bb73e6e30","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"bd2a5e7516f3cce1cc94f1513531e7d30b34fcc1c100acfef536a9c8f7dfc77d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"e8e5f262a2b19fcc4edf005dfdc5254946afa0a8155b550c18fb0e7a0228c420","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"cf24786bbabd74ca74daa883f74500d696a0a65238bb3b2f8245e381eee89ad7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"f72dc37b380e58b723ac76e959c2e12d3e1f3886be713adc598c82d4cc23e80d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"e7499926597e48da2526cf03a69191c62dc5ea620d8a3f63b26786e0506a1700","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"f7bbaf386870cec2af3cf3c05d8c489f32282414a8033ece880baf29f3b8c5ba","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"a1d302970e8c8fa9c07abe02adeb4b224f02d85e48895deb8f828eeff43f5a96","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"98c18ae34b7e378e1ce1fa0f00403fd7be064b1a193f93dc959bc487301c21f7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"98b1c7857693f4054accc5a6eed74df1493eae26a17ee749aac8b911aa2ad487","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"c9e837bca46bec4688f1cf857c9b4bc5c0e94dfc1d7599392ba8105191584971","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment.json","sha256":"8d2c1b774f144b862376ff455634b2b24d5b6932df98ce96e518929c2fbe4581","bytes":5255}},"requestSha256":"4254582b262874aa09fe871b837973c181d003d0e4088ea44e122a3815047e5f","metricsSha256":"a2932ee34672e52a7b2991d5dce6d7f852055876536ae925f34abf6beff5ae73","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2479261","recipe_identity":"13ae69530f08a05e0c545074d75d9660a704d0027a3806b2b14854190b28727a","loss":3.704439264163356,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"e1831b13f037e2edd427b2b0ba5e2d447dfed78810aec697fb4b91cc70a30baf","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"6d95b431aa55cf864bced4f6967f86d7fd92ada1979e12c3d80b5e2ca4f0d4f3","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"018f182be6d2a1cb174b9ad0e02262216e5c76c12510f7f4227241f7d0bdb7a5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"396d9481d37157197d50c9897c7eac46a43d774515e80dca4e69c52d1732157a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"85eccb29566e3b1532b05ecdcbb81889b2daf5947a39635a662b2cbfa1799894","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"9814cfc041d5fb680569bbb60bfb4fcf4cf7345df662f8d5c95f9f53c0d88915","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"9204f547b8337debfdb47e7129af2811d47499b93a67975f158670973b9d4adb","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"cc242eb06b5ea91dea1d8164fe03270276353c1febfe5a9cb4cffb06090a2073","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"45615c67d4ad5b1021179aa00cbeb07c88e6703a420ae246a962777e0cdbccc7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"30577ad2c88eebf203bd3128efa64bf4562f9335f20c6f544174d4d24f018a8c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"4c9a064a08b107983f189d72b426b4202010d2bc871f14c000b977840f39247d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"078f9d3b516be081670b90c48ea011529f3a35b38e4ec66f81a2268c439d6b43","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"ad33600e34b7b12cab3930bf41e57fea026efc4555182eba2bd6f7848934db6e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"2ac48ab25470b3e004cf547d46bd5c1f7c095e5b81bc51d3d4eddb4137c39f4f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"e0af8f85baf8c7e3e46acbfdc8acc140de0ce5e747d0a8bc4a322a96ae5d29bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"3b6c99cd04bf37f0306c6f2a3c981d5e110a5a71fc7bc82dc42b65d882e59b1a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"dcb042a3cf3ec740417628cbf07d9e9624c206a40949db0dbf8c412af2b94bad","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"3a8f45cb99aeaab604ebd57485c796c5377c37e95d583970f1271f11aff94257","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"acaf214d7ea672ad50a444233ee2268627bbfc6c86a4b60de04e1489b7e896a6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"3363f88541a3b1f19d0622a1dc7124ea82f851ea32b3b857d8d61894bcf3f16f","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment.json","sha256":"d89de5aa20db22d4dfa0034bd041440ab2863096a00f8683bf4b9e78ee38dc84","bytes":5312}},"requestSha256":"bcbfc0f87d7ffb47fc4f7ee56019e3f1d79f2d53f35730e274eb017e1d1316d7","metricsSha256":"63d7a7b44eb49ae618189f82d50ba804aa28b0f495c2760acdee6069c48d661d","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2479262","recipe_identity":"dd8f643ce2d9f90cbbf8fa069747b24192f156edd67b6c1fd7ae65a478f2fc7c","loss":3.718250021278996,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"69b9bedcd79596663d5592de6291cac496fa51e5c31f969317d8f497dcc27b2b","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"77c711ffc1b3b328bf4c06e8cda5bf6d2d7605f0a7b2749a251c4d83fefe2498","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"5c5cb234083799e870a44f34a3dcce0ebb6cba66b421f1124b7ff2802ffd1ca4","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"45acd5867c3acfae7a429b23b84914d9c49152efa2c8cac52c99f37b41d50b03","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"c6ae1e0e31834b56768ed390a2151e4e84d35c76f335c03129913dfc012bb093","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"16c08fd00ea0f0a488dec3a61b9c40fc513236f043abf2243ae537af5e90afa7","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"7740a95e27bdcbc28007191a26a375d30d6f635d9bf6f53229acb495c2cfa6ee","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"da78e25395eeb4b47ddd51e02eabc431ba08ab3762f22ffe9d3366ff56d09983","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"b280c02e6229d44a74f2e8b6c592a385ea846be030c87b143984b0303086f728","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"961e488a6abec901e08a2f4b633497d8f8b9443c67345cd005ca79dc35e43a8e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"90c922da88222d7125ba03b07b1c3c596dc356fabdb74313153663cb824c7567","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"19c50b240845c74789fe8ebe6d136c17d106c2fb5178caae1c79a96aad4031f2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"5f602604315980e38cc65fe66bcaba4913bd88e31c03d9800703b9d2bd713bbd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"3a53da8d08ff647685e06acb5d32f4aec84f17aed02495ed6858a067b6543e9a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"9f281473c1109489c108357ff365da785b738453145ba4e4ba617b090fac35a8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"c92cfa1b20e0956e05ef195e98c2cc464afb0699cfecf3f80e175e5054ce4f96","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"356841ee229195a4162ce724c6dce40cb36cb4120d66b51ca1cf135ae16a431c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"8c03289c79fdf16dd833321e347991aae7e9510c44940488523601b515d9f373","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"39c16c29c2a3a22c3949aef11d3872798648799036c1757d62bcbdf005a38bbf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"ad2388e57666bb64ad059bdd26d745c9fede11d26d99182e90dac1b642778413","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment.json","sha256":"f1833bcc461e843fd172b31e55034cd4d1a5730fcd0e1c89c36c183e55bfb8f7","bytes":5341}},"requestSha256":"196a5f297738b6f98a7f93daf507c9131e9abc41bea715853fd35112bec965ee","metricsSha256":"00e5d7f628a65c3988dab0277dbead5deca68fb0dff89a7bf66789f0dbf2cf6a","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.01-s42","horizon":20,"attempt":1,"job_id":"2479263","recipe_identity":"fcaa4f2b5fc5e8a577ceaee8cf8ceca4ddca0ee685e421d0f5447acfc1d1a680","loss":3.6172776136154665,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/result.json","sha256":"bdc4a6fdb9b558d4ecc8b1c8032d037950c0c7c1e0d474ce436e55e881d54c57","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/final.pt","sha256":"658618dea5a211f7321aef41a49445dce89a948dbfe78d335cf81e24a7411040","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/final.safetensors","sha256":"401cdbb1147c4d8b0fa03b38316c802d36bfa4a04ab42421079052201c695809","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-000/final.safetensors","sha256":"21168d0e6e7f52bdaec71ea9abdfaa370575035e9f5b15ca4aabb1dc62193002","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-001/final.safetensors","sha256":"8f49a895b0212642ae973ea1bb5a25fcf0e6bf2411822c0db7653ffbaf7c7d6e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-002/final.safetensors","sha256":"614b01923fc35df6c304d863e52f2dbe2c6ac1dcba2143f102d1f85fcd483b16","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-003/final.safetensors","sha256":"7ca0fbef4bbab9a1f74cb370e751763761f7fd314cbd34d65691316e4e75835e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-004/final.safetensors","sha256":"ccb4b76ae827b4aedbf0257631ce58abb2e8903860f29ccba803b2f3226f1c25","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-005/final.safetensors","sha256":"d91fecd1146956ba22f217729b4612f12e28f7fc870e7175a1d10a39ae3a8e70","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-006/final.safetensors","sha256":"4eacef2e730f7769ec172e4e4915df3257fde38bd5e09d8e4525056155ac5162","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-007/final.safetensors","sha256":"1c5e6e664857a7fc6aa02076f66e113bb1dd855204694738683a654a2020c22d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/epoch-035.pt","sha256":"35912e0b633846f35866d0a56429810ae272cf73a3c596848a720cb72652ee59","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","sha256":"5547993a09998bf90e0c173a461125b091b2d970a1d7b2f4a39c22cd7b17ca26","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","sha256":"4ab2f07ca8fe22bccbe93f494451eddd39066f1a93810e47aff8a9ecd8bd475a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","sha256":"171e16af04812123e08390668edf71ed8edaca69c226dcba64d465f2a85bee47","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","sha256":"4870efe53e7b89727c0ce2c41b933c935e991043a98a0307feb193f5fbb89f55","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","sha256":"aacb89e57d7202cbaa2988df39ed53bc6d8da66b1d798c1c412c5e91217f5fdd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","sha256":"2e5b4552539e0f734ce00752309b1da013891e530a5927fccdb12c41426482da","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","sha256":"cceea1293051fb1bb18937508dc373ac0976ce5969fe1a73fa3669a4fe82a1d3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","sha256":"07350b674684dcf9d5d08fcb99c21e9eb7dfa6582adbee4cfddfaab7bec7764c","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s42/attempt-001/environment.json","sha256":"ad9a5c7e0c231c5f67fd6cfde44bb610d83c6608e8c338080cd059c9333f9dd6","bytes":5341}},"requestSha256":"09f7627b3b62c0b7edc6c4902953dcd228f18964d4fbe343110e7208b9e8f3b6","metricsSha256":"ff25697f93a7d8cc5416e875373daa86cc98bf334bce32a4bcad216d342834a9","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.01-s43","horizon":20,"attempt":1,"job_id":"2479264","recipe_identity":"8336918481f3aef1505c647049f6cfb40d6680e510f853650117f485ce606ff3","loss":3.6107255052433764,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/result.json","sha256":"cb574d13ae162dd682ff157c4e8749f82d1b712c532f4f454e5c65cd2dbfa81a","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/final.pt","sha256":"0cb735c814ba8b40ef7d21f1d1f48734f367fe131c19e36f2f8414bd660a4d01","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/final.safetensors","sha256":"51934367cbcc27efb14eadb6e8b43d711f5f8a954a24b1ee7434b1cd0b648c6f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-000/final.safetensors","sha256":"eba66787d664f273fb9cf091cb70ff395a06b0eb521e6c9f11179052c04e4f3c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-001/final.safetensors","sha256":"84d2d807c8eb5af0538446ace68f9044f55fe44195c6d1dfdf7f4c9657c7ca23","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-002/final.safetensors","sha256":"74e92c83f6ca333d4fafad72596e966b3f2c995ca351e82c03cf212a1fab4e26","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-003/final.safetensors","sha256":"dfb74bf44fe34bcfc8a482fc92a6f3258b865f1fd1a2dc2b0be6dea53c066822","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-004/final.safetensors","sha256":"db6fa130f0099d3e0229e80f0922b2f15ed5361dbad2adc68ac306373994d1cf","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-005/final.safetensors","sha256":"4f79cbd2fb42f2a2a6000ef7a8795cb6b978e68b9c222b2af2da583b964d7eb9","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-006/final.safetensors","sha256":"c7610f589b0a7cf45fd310227be9f096f332f0b66626fa1f731a70ce4eb90abe","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-007/final.safetensors","sha256":"a8a0ccd3926b650b43fd98ede7d3f1c39ee454cccfc09a70fe6c1fa4f18fd8c3","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/epoch-035.pt","sha256":"e756388d2804340f6c85fd02143dfee37497b8e30028d442167c171264fddf66","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","sha256":"d535bd5ad6ffa2f6a4ffeef1acefa3062507a5820d7bea7f4985fc1aba13d648","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","sha256":"3e0b16e5b1ffe3726a482149f80c2253ec6a5bca4a3258e26e9f3fa6bf6d7667","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","sha256":"110e936c3201d83524a4e1677215bd7a5fefe44cf8b680158689af3dfae2294f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","sha256":"5796d33563dfab20a6b4acab8956ea65d9be44ae9cfb83bb2edd86ed68b792a4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","sha256":"f996b46f5e3184a72f49c28732c10b2036dd8f16671ba7cf0fb39914eca48a4e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","sha256":"dbd96a845cde31867ad951565251c81bda98b64063549bf92420019d7a91543d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","sha256":"3428361fe6887b4f340fdda9c2c2d70900f59c91f0dce3092edc845640627eb6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","sha256":"4a152f9b246468f008511372e9c2c89389dc635ae7602b952c905e2d7c929ae6","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s43/attempt-001/environment.json","sha256":"ada813c22c2fe7dc49f17fbbe714c7cbf47b3d8d995ddee3b8c85a0d072671f7","bytes":5341}},"requestSha256":"d9661a44ef5ea997eb4cc4bd02ff78b346d43218636c267a30482ab6cadb1c64","metricsSha256":"b73c3a795811d8160d0ab38478463f3f471a0ec054f11576ee2f9038fb6ea804","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.01-s44","horizon":20,"attempt":1,"job_id":"2479265","recipe_identity":"bab7c9101bc738eb149316dd500ace5d6c2fc1d28c4472aa7aec21cf671f2529","loss":3.621138881242901,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/result.json","sha256":"10726ec84584b9ab2ff3600e2934a04cce0b0f3c4ead9f58bce42f6916b8c1c0","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/final.pt","sha256":"0076c916d60c89bb553a3e674bcbfa01e4f697d2197b2d44b6d8774870adab58","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/final.safetensors","sha256":"067aaa2842831e73b9b442f2e588cf7985b121277a7c9b69137b2ccefa7f813b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-000/final.safetensors","sha256":"2eea3ecd0ef4200a5ff88e7bfaf8b93f7f15102cf58109d0d3a2267c7e7989ab","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-001/final.safetensors","sha256":"7223f8fb18e39dba03ead2f628954a8bf439a88954f8763ce3c9e0901f66cf41","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-002/final.safetensors","sha256":"fcaab6a2ee6e40a4e2b56be09dc5818b0982407a2aae6a05227e0118d1e4b9f2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-003/final.safetensors","sha256":"406de6dd08a65ef8833879f1694bafa658b35d1485416d542eb9610ea49c3f21","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-004/final.safetensors","sha256":"b7a5e1082162041da2c2ff6bf5be9c06992e085c47b0b40bd2c149a817aec5ac","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-005/final.safetensors","sha256":"1b2302914b37e4db3835df2b3fd3f640b316326b2c2cc42b443e19f10354f85b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-006/final.safetensors","sha256":"3e062f840f3818bd6b69b5cd199e0068833ba6bb6f184ab433a408e1404969a7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-007/final.safetensors","sha256":"efb9ac01336eadab898e25de7a96e8f607e4124b271154e98540ce283895d510","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/epoch-035.pt","sha256":"f83da49806aa821c73ce53831a2665155af24fe77f082d1e6c1a13c47e35b247","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","sha256":"37710c837050648ebee9f87482bfe01e15ee615bf4d95333d1d4726d597e6b33","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","sha256":"4457ad64089077b5f60e4ccce1d09aed6551ea08a6ed1e7d375f3d7cb65b7d5b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","sha256":"37ec0d4cdc3c4b5d4ebe605bbf37574eb5bb5aebd467f1b53e54d3f15989b043","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","sha256":"155c33ee3f9bd8bf2288136ed3cca070a73233110030c2a0dd44b26a63f4a6e4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","sha256":"eb583717cf1839ea62f2b70e6a1789c91fefbd05dc8bcaf98c7221386376f465","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","sha256":"afa6f269bf11112c464b0c8d9b29082212fc24c79b8c6bcb7608fbef0bbdff76","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","sha256":"81808db32aeadfcd7ba1fe64d2565c9351f032599923f7448f666e8242adba83","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","sha256":"3afd4b8090a552c05f3126bc9dce58a9d238dd2eea24d8c43fce136bdea6bd5b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.01-s44/attempt-001/environment.json","sha256":"00e2cfb52cf9877cd78278f0c9666a8685e930d158f6b47bbbc123c8823425f0","bytes":5255}},"requestSha256":"757d65c2116ef3e040af5cd8a15616fb6862b654fd2f4c1b5c12963ea465e839","metricsSha256":"986fcf1cbb142a94879420b05d668ca93523a2a829daec95d8ebcd1784207c64","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2479266","recipe_identity":"ca8743c63b8ce2e52488a209e6ce1f60eb4313311e7366fb203225e697bc1cdf","loss":3.616572170045111,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/result.json","sha256":"bc9e21aea6cae87689f86c05c9ef0bbe27a78644dab16680cdb7d60f509c697c","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/final.pt","sha256":"d549daaa814c4c8fc5c02643e2557cd5ac260ac3df92b704525f45332f62ea9c","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/final.safetensors","sha256":"653c798720b47e15f559d7c9facc806eacb65171073e82f5725990c20fa44717","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-000/final.safetensors","sha256":"1fb3d0799779aa4cdce4404d3558b1ed8a56eb3063b11da094bfc7b4faa693b8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-001/final.safetensors","sha256":"1deb7349b9f402c2468219e7d77cddd265b38e3a96c39538c7e3bc86828a4665","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-002/final.safetensors","sha256":"d58c039152d7c3390a6d20bd15a673ac5c0387822536c65d041139a6e67e6647","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-003/final.safetensors","sha256":"27198a99d966ed02359ed7828d1d7c73fafc17df05b70d2c8a9aa6a6d5c59abe","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-004/final.safetensors","sha256":"370ac00948cde4d1df867c3390634e41166a4603dd71020729ab15694ffb8341","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-005/final.safetensors","sha256":"d866c5a555b15acde914cf2f4d2d57e0e03da6d110746a2250e9380d12b1148d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-006/final.safetensors","sha256":"b32e7d98862ee7d3816fdd3f4f79edcfa9750b00c2a4e7bb48c9e3e41acd8d68","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-007/final.safetensors","sha256":"3708265b0552a3d614a87b2870cb953e9a1ec69886d78fc322de3a2de6fbcde1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"b1b12a55a1374c3c7733b48be7c92a39fc96c11dd637a3d4083cae5d6ffcb64d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","sha256":"af61e0e7a2b9c35fdf9e1e1095daaf79c7dff71db5644ef5140b7c41e5092da4","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","sha256":"5646a0f8475dabff8f220658b70ba18175963b709eec88b547a332ca595d7d1b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","sha256":"41ee9bb0d1bdc1d41e13d92c714e5320169006c94f8cf9e4155de2f907f32212","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","sha256":"83a21bfe191ac52a4929319aaa149661313a151ad72df0e23a4d0948970b39aa","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","sha256":"3afcd134abe4cf6cb71692af7dcbd6764b63d2b0e75899fdfcfc2ba3b5ccff5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","sha256":"c0635ad9e7579e3a68544639326aa1214e7567df4cb2370449ae86235bf820eb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","sha256":"ff61bf47d5485e8dc960642927c103dc3d5ca3ba91ccf5ff4513746fc1d3e5a6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","sha256":"1b3bf91a744c0b93f62bdb71e94e58502fe0eb357e7009d3e2b22a5c3911141e","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s42/attempt-001/environment.json","sha256":"7a967948778a8cf2d209f5fae9b1b84e260b50603731ff0e36efd07ed41f4fe8","bytes":5312}},"requestSha256":"bdb2801a2b54f919b5e3b29dc3c0092c82e77503ecc06f8583314dcaad0f52f4","metricsSha256":"a203c771b62e13844874d3482b0c72e7a60f12c8eba9fa38e56c97f7542dce8f","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2479267","recipe_identity":"75c9eba9cfa850e9b1ca1cde2336270bad2041cd056fc483e4e74565809cd587","loss":3.6115444370452328,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/result.json","sha256":"5c81298ecf1353edd2198945f36c557131886d519e9bc4f3da4b3f6eea124950","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/final.pt","sha256":"6065c83eee197c490f1a45e03a5fac9327d61647682a810652d519e428f60833","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/final.safetensors","sha256":"6f9603ac80a0a8b506cf4273da49a883461c62b9ec2e9886c1253dd30aa822a1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-000/final.safetensors","sha256":"9a3e9a1818387c22e3a2f656f13bbbb435d5d81af8df85d15110a5c72172b866","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-001/final.safetensors","sha256":"7c0d832bad2bc22f0fd48a55bf261736410202980a8c52bfc8e705079e7f330f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-002/final.safetensors","sha256":"49dee3fc4576aabbdd680a9cfcaf5e13d439d1eb0dba961c1ccc4a0da2cfa617","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-003/final.safetensors","sha256":"360a48cd0b86f0fc654bb2285f272d9f75e3ca16d65b63b264041ea79abf828e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-004/final.safetensors","sha256":"a06afcf84f38f0a34c0d7ecbad24a4c778a5e4d68e1491fe56afe8dcb939eb2f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-005/final.safetensors","sha256":"3d8c206e455c3d58a8456d9e2a15b5e6a2adf03f376054e9218cc3a2baa9c46a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-006/final.safetensors","sha256":"aae24d06f56a7daed24fad26eb5b389356ca065095b653706f93e53abf996ed6","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-007/final.safetensors","sha256":"15dcebe74c84abb297a40b3dfa0d85dceaa06a09bf5a68e0721304bf9afe3521","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"ab68063d52ca46fd881636633f2fe851d0c32b2ac9ad79d807116b4330257f72","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","sha256":"352bf877c56a176496a3434902661763520de0166315a5f2ee5daca89c8f1b26","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","sha256":"cd3f4c2bbba63a8ee5d73773c8e23261935ec1b1a50f8c477b4e7563f59b6fa6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","sha256":"2e64568e6c5a9771f200266917fd8bcc6781a28e5b8db539b391e343d3449af4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","sha256":"d037d7be5cffe9647b958343df5122d19d5ef4403f5118f0b7ab08b5bf6058c6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","sha256":"53fde4bca96b100faf79d237e304adffee7d0f4839753030fd9027bfca3b3838","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","sha256":"fae67da23339b67f84fc3258db0b74414634d589863270e2d29890631098d4a6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","sha256":"2f0dfffc5ea5fcde9b48d02ecb187a1705c36ef752293d0ecc55f00932123ca4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","sha256":"11d68cde73f9fa07e7cd3e82e26ac5d91248f9ca006118d3535976dc30012aff","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s43/attempt-001/environment.json","sha256":"d58ad858644fc7796eabf0395fffa3b457550d4ff0abd8725833199e8623b9d4","bytes":5341}},"requestSha256":"975a9a5004391f23b0f314817dc5616e038007697c20ebf7ec8d1980b0da00ac","metricsSha256":"c7072497d70fdfe70cc7cdc009bc3c223c4521888b0a64a4a301a4f41a857c53","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2479268","recipe_identity":"ed9ff209fdebf9687610c7465eb547684b7aa1bf5887e62a2d3f3a594bbca199","loss":3.6192615827092975,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/result.json","sha256":"8b26d9a3d22e8f0bcae726e6c9b0fb6ca781df9cb53b475d1171ff2544d0e227","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/final.pt","sha256":"176a575074bf9cd67387795f2e28f3b082acae57c50b2ffbd88a63c4b27e4f4d","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/final.safetensors","sha256":"65494ed697a11b1244cb1b1d7aad42259e6127a57c7eb5abe071049d356965b5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-000/final.safetensors","sha256":"b395dd60b8bcedece1d30e38382319af14125505bf11b30c352fd0175598ccd5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-001/final.safetensors","sha256":"e1efc4ad4a8923a89ee250c02b0ba0467d28e0eaa185680aa1e7edc2ad4464e5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-002/final.safetensors","sha256":"18fa7086b39482ee9a48f49ec71bc7baf3cc0003f9cfa220f03a60a870867dc8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-003/final.safetensors","sha256":"9c8c396799ff590066fdda05eed3f5284c28966a831a0b10efc7c0ae7bc1a418","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-004/final.safetensors","sha256":"502db6954cfc1292cedbec2e3ec74b2d885950d05a3084dc4c9d287247aaa336","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-005/final.safetensors","sha256":"660dbd6d0598986bf6f731805a027849c5f04edd5066e18ca3d59448beb560a4","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-006/final.safetensors","sha256":"367dc1caaa5171a5dc03a7e2ae6e326eb4961550871f8a7c774b725f79a9034a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-007/final.safetensors","sha256":"feb438d6079fa1e1b8f85c76b165e661acb8af0b04367b6abe70bbe70df482c6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"53d24926c435117cdad66f145ac21dd0c87ebf0b3c528c12ee11d1926537bf98","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","sha256":"16fc71bc42ef6c139373ab741be351b81fe5ba1e9f4fde2dde25bd94545ddf1c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","sha256":"cad595bf37c7e4cc5ea493fdc90e1ecb12db5e40d6b643747e4490892cd238d4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","sha256":"b79b2a0f460a4b048da02a191bc6f09aa361a51a436b8332f7a5c02a70cbef85","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","sha256":"c5fa6e575d4fe7a8a41f87f06721ce36242e334940f680bcadb0c8df4db3a077","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","sha256":"f0970114b6b3b3d4ed6f82c0f151f20221a77d6ecd08a432758727f3e06ed356","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","sha256":"5508e192275580493b74d30d10a6015937101fc731417bce995280a84aa6100b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","sha256":"22fc4e94f035c9e82247027ab8085c53b8936f2686197285dabb6b7eb91de82f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","sha256":"cad6c7e819788e0f87e9f3547b29dde3013d6e092eab32a07b7ca238c8fede1f","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.008-s44/attempt-001/environment.json","sha256":"50117fbc2467a28217e51340a883e826cde2f825f07cbd44853e115514272711","bytes":5255}},"requestSha256":"28c8d43cc67fe71a248ccc16ba9dbfa6c72b5817aaaec45a32752b81a356de1a","metricsSha256":"93d90f6c4460e2fdcdae9b259373397cac4a52d2722c28d33aa0cfc043601a93","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2479269","recipe_identity":"bdaeb31ffed42734d1a9e48c48f46aac1d6b892f6893e8337a9d7f49a2b839c5","loss":3.620306398628966,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"15a54be093a272c44c2dea0f232aa9669ffe4d2a89a9e1a4be62cdbf4ece00a8","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"18c06bfc208165effedd534b32b809c6d8e6e1ef503cdeae8e01e2e42d818fe3","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"e0b9da0fb0a446d7e9ff7d5ec4bc6fda735f99e6b288d837a812287c858b7a67","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"23e8cd611c8d0b7629d16ed93c1647a7d646e00b35634233122e1e71cf350a05","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"08d4db35851397e5558c30d9e3ebe0b15b6b39e99f41d725cfc755825018a265","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"6a7562dc75bde907fdab7676f6ddc06f11253eb2ce3a5f453d7142cab8e99044","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"02ce64da46349a0b880b17c3affbdc0dcc0b64146326f3507d6756faad33bce3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"616d54d6be9b4fa36cee3e0c2b497269df6c3759eb2f7c4ff4d4c183de1a8b22","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"65e8f7eef3048f1fdd464a8af95fffc14697649e67bab4168f42ba866716cf28","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"3febcce307f909b746fb922578de63713932c1049acbd913026be985fa847a99","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"329f494c3effc33be5bb144094a496758e92c753d0e9aa0deb93d8be1d3f4390","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"b670d47f40ded659e8514c6cdd9dd45d8ce74919a507710810414dee826cc2ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"f725c996d5b1a4a3d1513e34d1181ac1d6926c5e0c972a3431b91c788e0626bc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"127e3a37f8a2b79ae646251b66556820d4d9f13ea6288bfae8706acbeab9be2e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"b46f1ac59e0be2e0ffb3b4ff6c2acf566d7ff8182566a5bff5f7454dc6ce7078","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"5002f90001ff4265f96c9f3c4de82cf21eac9ae165687366a68a5d7872a85050","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"734e1f0c8de703414c415f348e12a570068338473043d032961c6ef7152cf422","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"70d01bf9df2e35b008795a7c1ba80d2db5f7edd547b28f6f4bd23fbed22cbde8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"eb1bb4c12ec92d7514762de1bdcb059e0e766f4c371650b7efb2db7b73f84060","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"e0906b1355c37b91f7efa38e095bf45ce99f38182966de1b71bf02301a9c88ff","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/environment.json","sha256":"f494339c7177d0040cdabc6842e2d3bbe0676b6a319f7b9a27d70bf264214584","bytes":5341}},"requestSha256":"8c1f959d478dfa4685951a89095c9c1f04faea468054f7dd134298243159e591","metricsSha256":"c9e913c8e5a49bf26df08313add426faa0b7e9406e9a20eded17aa28d9fb17f6","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2479270","recipe_identity":"7f04c528708390400b71e322377f8d9f217433c9bd045bd7728148241154f57e","loss":3.6155477850644306,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"98a6afd377a438c608e295f40f2a3db311241747186de3f5d3005b098933afe5","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"1c0332d731718e1b19dc0e7ce64cc977311273dfa42efe1e69897b741e366737","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"d4402af998035a60dd1c77867a729d6a66bc2ac5a43c18d4b4e67c8e97c4bf10","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"3fd3c88459a7739c26a28abf82271e1e1e8917410b7435a32742832e1a61d9c5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"430541bbdcdff7a3276748fba197d8dd97e4930c417ac9d8aa2530489037ca2a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"bcd4bdf0b9d743373afc8290c73fdd7d4506597771f5853c7869c3ac14e33a51","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"53964d13bdf74c3cd54161d83e88626849a48f9d0be201c596aca3b4e425ff4a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"1008418b66a8e071803a34f0e673b0e17674506b8273521f0e498a2bb882f314","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"f9f6165c09a9327f6c5388f6dbbb2fc5583ec19fe23ce8a7ca407c6fde76906a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"b443de667771da9c845663be503f904fa65cccec88af889c5200467f7a5399c5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"3b6ce1880a1d9726cb314ee53332d65b6bb84e1416a9e563df156d95332eb11e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"923b573808ed7046d068aa4caab530a26fcef6dfa047eec6e306523c8c11e280","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"6e8ac07e4946d2859825479876a321f5a529eabde5b1bc872bc5953b1bf51c7f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"7ed3d7a7bcbdcb78bce89bd772b600e7c8b7f476021566bf1b5bcc2eff18e11e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"5ff09b62ccc3ea116c1be90123aa8d99ee1241c105b64fecc482f24216737f80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"2a72d97d450c473539d4a5033b43fc0754494db3292c30502dbeee6057e93304","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"2f5bb962c0d6c0edaf7f3371266aac3537acdd00689b5e776507b6d78748a045","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"623d6b661ef710da2b1e58f87230972d4c485c08947c616dbcc4aff03011e28d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"bd763b6b4cbe935946825e1166da362ed2f16934be4fc7e3220e070fe0c9f182","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"541b3caff8ca603420d810ef4a2fd58d54d62bbaca4e29edc38d031b3d20f1f9","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/environment.json","sha256":"e002bb9c37407ab673083fa2e3c807fac7ed880608076191a738fb004bf5a548","bytes":5311}},"requestSha256":"3c9f8c08c51f58d3d3da8fb3a6f3e740e27bf8ff90cf977a10bc24cc8e783ff2","metricsSha256":"de4c63560023849a6b3bef0ffda387db9a5d439bfc46b6de516831f86bb64016","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2479271","recipe_identity":"5bdbb9c80974b08ba729a57cdff4f03b58be4dcc61e81aa45775d48dc588efd1","loss":3.6195543470806535,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"0870a90234757241e5a2c251cd4b2cbc247e0640ef9d7479be95cf6b95236802","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"8b3ab11fd832cbee0081a4ccca8af93efe532f2b4744935ee872373ec3fecbbe","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"0e51bda883db9a1b9d14131265fbd69f88e364e76f246c4c7cdc002908a5af3f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"48705ae7477f5b72798632713e99808f6c2bbffc44ca241e6a9ed819d2aea4bc","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"8b679f31b6a1d96c8f1180144b19ebe2d7654676ff12e83a22a357bcb6a0131d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"cc2fb2fa0bfd6131818020dbb6090a9a0ffbc68ef4f0b6da34892e49dc672ef0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"2b116801beb56be00c426d8480585fbe93412a46c68a08c2ae3548d425cc9bdd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"6ba4ed32f7cfe606d84f4a31882f7332d5421573a81e2b631a54905378136a45","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"dada4974e14a61940518d3854aa6795300f654a2525338614c7843d0b232768a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"edcff9d69f2319780910588cf2eae7acd489ded38769107f20f78fadb00db3be","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"225e38e997ba19acef54bea57df66872e7e90d0cb8556545b907420cb6037738","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"fcf28f4e72a1ac61872ac5032dafbaff2ddf1ade36be44e621105686d4042f4e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"3d305466dda96952466bd5ff3d6b3b7dfb2363aafd874bfae8f17f3e1df46f63","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"4043971df8ec44df7732211c1b2bc91805b998d811ecec03e1bce460464b7797","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"0e08b6d57455c72e50b721daa04ceae00456a69e8d9199beddc0b747cbe08a54","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"91f87f62a16d074c2c3f6302aced938df99fd3718792260029c4a07b7d612998","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc9147e22a451402a7b381ae1dbc5d10bd07901a5b36953e1d2a595121c4d641","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"5e16ebaba039c8ded44c259a543a767f652e89e107c1f8e487e0a81bee413a0f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"199bec3b1c16c0e71b658f5cd9c39aadb5411d1f74787bf08165038a94efdb1a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"4b7760e4bff5d42ada8a00058423b06eaf46a10574bb979673a9f38081060ff4","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/environment.json","sha256":"0e53b754fada1c8fcf45594733ae286e8530869512301e47609e4a6ed3de9386","bytes":5340}},"requestSha256":"37523218fcdf4860b73ad2658bd0f4bff012ab76246fddb7fccf2d08b844ce38","metricsSha256":"676be6cc740ef06189e8ec48b8a935eec9d9333926cd65c917653adf2bff4eff","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2479272","recipe_identity":"f7163331ba61f1e9c7934a822571d34e39a5d017d29de9a7f48ffde9c0883f56","loss":3.622101986832102,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"65d8d799b62be6fa6862877bd6a42e27b557e43d0a836d396a1fbb1dbf1e4a32","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"a41cd7ce415a71f036bb30059a7968d415772ebf523d167e4657f9ce34eaf7e1","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"898f2df777025abb709c3d55e8117e042ea2b8485f12104dde02e85232f92812","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"ce5e161e612b05ecadde723dcda2ec680bcbb80d59a000acb66e3989a56f928e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"9eb8edc40690d1647992b1b2b4148072710d440e42069ab5045d92696d57d3b2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"037c700a43cb48378f7e3f68dede2d23949672d120c1553e5c4971f4720a9bab","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"2aeb7a664e7fe849ceb7f09b0e2c278bb4cb452ed419e5cf67671ad19e7e55f0","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"5f32996b22bfa8dd7779578be0c4d754608441dd386755a4dd247d559282429b","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"02d9e08632b31c5267420a71161d503f4a63be93e4e57ef44f975cdad51031fd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"3949006cacd86fa1f8575fcfa270aeddb7542bc1654089570287480f33064dbd","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"a9ea5662e13d22456aa9bd87e8527b69ba9abc299102a2bf2f15aed260ccfc1a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"e7755bbbeb7bf7853a20e381cbc60c69ffb084e5ce50580adeaffb14fecba5fc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"1a9a73429c58d56eb0deee75af7cf7d1af4f37e6492fac2ec09c239e0521d358","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"cfe47560b7a5036b43d073e2568ef04858a8dea95c52d0bcb43a93bcd7f1720b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"27000aeeaa975e130f135ec4afff230e0f9324f7dc411fa9fe18aedf14700c62","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"230b156cfe82e3e3d95c3e31b850d4b537af845501d757fbf7e95d0450174b0f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"00e1e3a835824cb693f871b405d1e17c29dd87116cda2833387dec0b7f98e513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"1f33100c27256cbe32e3c3f4775c87396f29aa4aa4cd39b71026f1fae3ebd9c2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"ea610953962ebfdf9c6be8fc19e0fd8b1e7c6850e6e9778e4d153c34c987d115","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"d8c3440fbb761dbbb6b7a8d96721e3714ed4c35b8750ed4170cc7f7b3ed42eed","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/environment.json","sha256":"5b31a91d61bf546b471a7168e03588354289632eb53e66955161d43dfe6e7f46","bytes":5255}},"requestSha256":"0a756e16f8e53c196887c7d5c71be4a8e9a884a2cd2c81efee3a921d9452faa0","metricsSha256":"9699f0798a29ae6bac820b8b7cc9ea9c64b4ca438c48a95d09f8f054cb21d90a","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2479273","recipe_identity":"69d46d065b831692a48e7a7e44b15086074f96d72203ed94d893bed0a3a67224","loss":3.6267173488083095,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"a0a1dbcd31728cc73a6997d6f1f9829ac4fd10c37a1360881fcc51daec0b9744","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"a50c8d363ad995e496e9f421e40b8d5fa4960709ccf5f45e47404e2dca51a3ee","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"7bb633d8e6ff24067e5e3e7fa52527737f6c3312656d3c65cfe4e39b9d6eaaba","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"7ebda58c0197ca1ebfc0ebde28a87139ea904a5e4d71afb5c603ee58401973bb","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"024940343105e55184143301e6ac6f4847de510121613fc9101ffca069ad8b59","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"bd6a3727d1b3cbebf4cd9efd2a5a2e88285f0f872c88e543f0f94528202a5256","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"cea8a94c5f153c8e7524416742889b0327a93380b9ee123541e6bf97d439a5ef","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"e4979253446b90ffb5542268453cc0474bb78a9ba58386fe328e65d99d92c254","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"b41285276bf54e21a59e264cd4057e55110ca53f22f3e6c162c4ec5c18b570f0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"7e185eda139dc5fdb31d65ac06ec5b1430e570196f963b44b07a53266fccce61","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"238d0bf2ac46e5187e0d0815f84490e7ddfc8e0e5e1d07595a77b6c23b80061d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"98b2854c5cbc83ccf9fe96f3bb14691f1f7a7e9592d0c1958518a99647ddf434","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"73e764f29ae5a903cb8307af2fa6a5e05c4766dc6db97c79aff94926f5eb16cb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"30fbd73361c80c3835a11855b69d0d7bc2cac4bc48c4b0677aac667ebca58c8f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"22593d8e25c6be6987efb62301e2ff5f3258799e00c84ce2603da9abf1d6918a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"abef92d6356f1f96566aa47a9fa60376118bb2981dd764425885d24b96f1f2cd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"28e86636bb177ea934b34751aa2e629b05e42d77bb4a71455a887f468ad9adb0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"965f006c5d0591a08532a1f0d141562c5dbe6c325ffe31d470340dad4b6bfa8e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"e6ff62ad33073d2139217d4d5ccbe26c2fa7f1fe6fd424d03bb161dd6acbfdfb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"4c11970f218cb8f63bfa85b623d530d2f63d73e45506d74a0d15207adaf8f2d4","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/environment.json","sha256":"8e2cb9edb80026c337cae7dff402272a0e1fbdec3d242bfd38fb4d37e587bb99","bytes":5253}},"requestSha256":"438ceb2af06f5dad5cede7f88a635ca2ba17193402ee951fb45158b222aae248","metricsSha256":"637c4a511a23de5792f7d5d87fcc419660e19574581529bfb1fb90df48345e4e","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2479274","recipe_identity":"8f414e39923d9737d9f61c46b9e7c53382a8504fa035cdaa39fdcce5de0ec68d","loss":3.628047020082648,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"e62cfac7237c8c09bd1f01a5537dd318b5fc2cde2153a699c032ce1b1ef381f6","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"cd58575d6041445b089bda49458194fba2d71a9f60efde00ddbb5acd30ece6b9","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"8d13791c6ccd17878120e5d30e19e0c27dabd776de19f1c4f01eef7ebd5e06aa","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"5638afad20b446e2019eff35d9482d278c4880f1bcef4341e4ea02d04f9b8326","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"8c66d7de24f9bac1400cdde2d0fae0a2cc9474808c2a6c4a6c2bf686563a0dde","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"0fed4d156576834fdb72457c2c721e5711f1c4b71d02beac9ff8989fa503848b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"937846bf92432d03df960f0de41282428af8981d721eec63527b10478c900752","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"7017e7976b310239e52cbc30cc136bca592227b293072a9bad0ac7add4da7feb","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"6777a98ab25433c52fa544b97401443049e856b3cc3b9b10dcc582a2412fd5c0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"d7e2a6f8ff688f33ae51bdcaffbfc5d8bb62de978f26f1940598e4567e693a17","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"039fa94521d96068d082461710e977ea1e56ad5c66cecc671c51eeacf2963b31","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"d171f71a8ec32621f633d46935a922ceb239470d18e058a2461dd18dff06ea99","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"225ba1bbfd0f6cd652bdf6cb97de2fb4a97ae086966d6102b129272e8b6f324e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"225ecb60bc21be28d5c55d7505e77edf3822f70c2b45a9cfe54c457b284ab2e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"7dc6f062da3aebbbb5b978675d0a50a1b27e6933fb565bbfb60f48d1e5975817","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"b6874a5120c1e6a59660b4b626614eef77dcc83ee0ca43aa8698872b6e269875","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"216c7fbfa1cb9856bd48304da72842253edc07ad71d3f404d02184a9100201d8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"102779ec2ec26459f3e21e4a63605174cbf8f3e1781652cee01cb58fb9e4eafa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"6f82357c83258af95145570ccafd37629986029324cf616bbcd013bd9daffb37","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"f5b6f80ad23f6bcc2c0a5a867431f7411ac93e4fdefa50bd5103fd25bf9d24ee","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/environment.json","sha256":"6bdfa1fcec44875fce3636d0d077da5c39ab26e9e9eaaeed51c2b6a9f7475d0e","bytes":5310}},"requestSha256":"86c9f4e02237a2c5a3e525344ff3dad1b298aa28ad378c7896d782bd4c64150e","metricsSha256":"ecd1401fea1caeeaf44555e82f3c207cc31a022be25e5fea27611aec8cd3a23f","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2479308","recipe_identity":"e3d02fd6cd61acba47adaf3e0c32f4f793383ec17128d711763b64a1ee20f8de","loss":3.6329989668651663,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"3aff46fbfd02ea7f227cc794f10a231ec02036bdd51c417733af4d8c5cab05fb","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"b1e7a5974e56df8360357e212c73c8c33527729daf5e72eb627dd41fcb4ed054","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"fb3efb808a2f1805eff085865ea6d8a05a496aaeb2857193a4c4234dba5f73af","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"548d1633808ffb92059475f1ffd7cdcdfd6c33a4b650820d429184dbf15b79ec","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"43a0313c0e80d21f5ff8861ab7c82ad8f0ce19fe44b7f44f9adf1c9712583d99","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"e169c7533a2f23833f95b8e87a5a0b98302cc2e5451c600824866c4b8aa7fa35","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"c3e4227921010ac1a3554b5860f4b32fe8b23c0960d1f45586f2c4c3f52ac8e2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"f529afaf1fbebe8cc2fb47050bf63e8c6ea693d7a81b615186efbf926dbb7ca2","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"2d6d5411854ad653eb2c58d55b1e03853ad1112356f89140e0e2a3795a254a46","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"034209dc14a97928a6cff979b8dd59f81068543bf83813c14da1c07f433aba06","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"78a8d69523d569b0ed10459f38e13d2a219998abbe02af5caf7f82a641da46a3","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"de1f3135e061c79b8c4f9ea97a08c701a0476d5c95e499b082d809ff1bc20bc7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"9f9830c8afc6a54d9beb0e676a9aec782cfff507b43bfaa3a821908552e843e1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"c1ee97dd236d6078fbd795dfbe2e1eabd03f497f280682a3c285b3a797be86ab","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"fa6b8ac52f917e00b9fd471586400eea88a5aa1b8796f5f8bdfceb6968b4ccb2","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"05310749a2307cdca43123d5151340850b3b1a497278db1ac4ff48f217091346","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"7cfd163fd4c883e094e76b2f1ad1a1b29974bab40d05306380e55a9211d8c955","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"3b6405d0b1c9c1c5517860739cdefbf84a3c4f63b4e455a3880ac8df91188fdc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"a3e806436b4b720c4e87d0154cf5764f653d4e1f9c670cc9420b26003e826efe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"b19838a68635c9458566cd56cc99e2b9fda327a33555a2b140a0b88c3ca6b9e3","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment.json","sha256":"3e7c4708d31ef9202e044a08a17f594cf4272d02003abe185e369e2c851d6e82","bytes":5312}},"requestSha256":"3782c767954d9fe0817a7138f54a3f4866fcde078576152efcb130055574f51a","metricsSha256":"c877fec6b6d48e43b30d8ec63aa2cf2fdeabd5aec182a3271c395c4f69ad33f1","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2479309","recipe_identity":"d3c1dbb8d488baea40bc613953ce010f2239aee209b001773709f1180442a548","loss":3.6417305349872295,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"635b15a36e991dd5100b5dc690bda22c2ca59db6e86208bf3bd319396c3038b2","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"5a07ca30ebfbd5f47ba7495fc8594452d9d032487a82d2143756c65385458fe2","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"eed8c920b6d45a47001f98ba81d9d96b8ac20253ced3091baa51ce3937565ac9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"9e4f7a9e6a44c8f8aa88c61d1b476f73492e80a1ad782cf76742d659fd6f01d5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"f2e2e1756eadf92942eec88bb77d0887c51eaf6a3614ad6994642f9caf338f09","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"34772dfb136957c2ecdeaab1bc64a54552937c8f66ab42f92a1457346ad4297e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"b51dd0e5c0f6f945a6f28b9cee744337a0208878e31c8f823f4b21aee5ffce01","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"492e28353ac542d94d4483539569b960c38714a0ac577843676f2c10c77a59a4","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"a6d11192b0bb451fa7bfaf344f25154791df15e0fcb384dd23a47fb2d385ed4b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"6fa3ce28a8ceeec5d45138162095f72dddaceadc2a52718e8c0339ddcc32c251","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"4d6549f02100ba1251ee664a4dccd29b8da2f57b35cfe9df0ca7e33fd71fa89f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"f96a925e396d52f48eec76aa9243262dc3bd797b3b5521b7be1e913dd0382bfe","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"683677a53d443729c216c6e38e005ea15ac47f1438ffcb3ddedfd6e1f818ac1e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"e2a4f6101492a09293d40f7795f4b43280398fcd4470a1940dfecd65a4456fcb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"c555a38eb813470486887c20b54e721398925611de6f259ac8e43b4b87cb5b09","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"fa581e8772e628f93ff158f8381749e245a5a01c5b6d4649bd768d4d33be50c0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"701399fe360c45a403b114914d5a856c7f2e6645eb1e6a7c8306a58d529aefcf","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"f503ef79f0880b2e283857a5fd2a87176bfa6c160c5d4d0e178cdb7a34c1f905","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"918cd115b511922cc333a2ce8cc68aaf74a0021f578f1dd7dc0833f404c054ca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"d5c2f28a77655beca057bc498d60e0eacb7c73ff9e8aeefcd685d9226cbb066d","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment.json","sha256":"17776edae2a01cda5df0cce550dc8afd232f0485c23bb5deb92355128df4711a","bytes":5341}},"requestSha256":"75eb74dbcdc35d7e4bfe0f0dabaae34390620851c7c4474b3cf707e65efb2634","metricsSha256":"371f91d01d28c896f9cde8926d97bed969d5d8eb8ef674bc49b15fa987e38d49","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2479310","recipe_identity":"58fc4b5a07c72065c270177ca9a8818e281fc8e54c6ac6e8798579b51cd74dbc","loss":3.6354018805587303,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"8de4ec6c7df7e103c288f7db12b375a93b20d86d42c27e505591f475583e4d61","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"84c6eec5a46dad73a66450cc213a16df7d55613d9d82dab6eff4c59ae13a25a7","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"464f376ecfbbb678700bd097abfd653e936e408cf476b0761b48d87f240d56c6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"95dd8ad75395d09932314a3ae891c08873a330ee74a448e4b800c62fc53d22aa","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"5b6285958c71ac4888966354be5c378f7460eb275d205e7773e0087cffb7a1d4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"c497a008f887a0cbddeba5bf73da5cfef5f7daa111914ebed1e38ed7bc0bf2fc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"e36b2fdd69eaa343c641b493e44eccc578b95d9f14d975662bac341d58f9a1a4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"d35dea3a106b07571c8d34b6b0973240b2b481d60dfc30c0a282da4a7a2c637d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"3fdaf8f9594d4dd71ffdd52ea803db821bb32b38d42c6abe27b63593d4c4d3ed","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"e8fb1d2a933e2b2eaabacaa860b60c9f1781edd3b4a690e75a595c3beedaf0c0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"da4936151028ac5e155937a94dcea866456f573a222af565d6dabf687bddd6ef","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"36d270ae712b4f2bc4884b719c13e091c7b9989d177445d99cf7f79e061ddf76","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"4ea3c2a9fd883cae5bcffb2d758ba66cb37e7cb863b787d05bc2cb15ebc21425","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"39f83641ab9e8ec5f8ed5d5506bb70bf718e7ee637a8bdb9b26b8bf323834f7d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"351cffdcd7dfc082fb2e153d0c1eef6ae7bd9f5a5136e4cdd0cbbd33167cdd3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"e79a8ce4eb95a4b6af2444330c7c9817a2b97d65905f55538c876f7c3a424f00","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"408b9b059b621c1493989b270170c0f419189b4a428407a3af30fe12f3048996","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"ad891148456a307b360a91f898f18ee0397d13e14e884c3617c83b0f5a48acfb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a95a8a706138cbe200b0be61432ce1fb6d7984057e31ddaffe2fd9f773435ba3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"c408f48a63923d5d1cafc71572fb3fa6e6eaf97ed80d21a265a820ea6c3427a6","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment.json","sha256":"61f603199a4153cde89bcedf565bc203dd7dd109ffb55b1521dadc69df7cad35","bytes":5253}},"requestSha256":"4b442a371c48bd0777514b38a58a1406fda1e5aa80110559d07ab00f2044b339","metricsSha256":"9dc893e7f907a25dd427af33e253a034385dc631b9b4f10e51c692fb92f6cd4d","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2479311","recipe_identity":"d28361c031e0ed7003897ec62a41523dc30d8f06ee73fa1cbb831f4c2cb8173a","loss":3.677922958460426,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"e39bca57b0b984028856da617a7702bd41fdd7a937a6ced7b5074d509fc3150f","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"ff451e993ec29414666d3522f97ddc95fd3e88d00f76808ca11ee8d361fc17c3","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"7e426b1174cf7b6b3870b9f653b096f01ac7dee635cba57a2494a010d89a82b6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"33be4b0c36c559080525348c7891d5e68de9ad8d638006d094570cd5fb21c1af","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"f01005d9a732500df43329f5c1dd01a839cff89b6b20b72e782eb1f6465d0565","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"faa7c8d930dd3fdae12df53b2fc6c39d9ae01f5af59c4b3200b4087c61f84ed0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"1606a44697773ff66450147d4a49a4166d1760f6dde34ee2a1e9908d626ec225","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"1f5edfe07dae10d8213a5fa9fa87bdea72914a9c56eeea0b49606e4f6a62b10f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"e71dd1fa0b163cf13001cb3e36a9f46c729a87258a3f304126be8649cfec6f4b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"dcbdd415aad66d407d00669d8c55ba0a12a29cedd241118140dadbf820f49cdf","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"a87826bcb894c329f712afda4cb08b56a82ac08c417e834a152a4d5a953ebb7e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"5a3b342a49884306f6975bdf929cadc7e8088eb289f82e9a6f9b7ef5edbff412","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"87bb9f88a9f997578492c794c8cd01b89462734583b61af9258b4c1583bb536a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"7acc77f8458dd0a0ed27489ac4888f44725b598619fe310b2871ba8326a66a7b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"350cc763fcaffb107c8ff5b0e011e0b1c01ecd060ed704dd73774016b3e5c785","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"5417724227b23e70d6b655e66c6d32a8ad112a437b16a413c255f2d2e324976f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"362701167afbee042473651d9c71ffa445d385c1990fddec9cd42fe751ef94d2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"db1ababf8c2a07cf7fefd2a2d20944f331e73afc52b5d3614eb49aff2397a719","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"ab993bad96554a0f2550f168a46ab90040ff9e48287265f59566b390263094fd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"79272d957aa2b498a5bae257478f324a2695a4b11ff0235f7c7fb6c4e6b84d70","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment.json","sha256":"86ced0127c37d246ab8f4313984c28997958625670ae47962ef9e4b36c0f4464","bytes":5310}},"requestSha256":"305e89c4173b248c4887a3ca851d7161a1f1fe6d8527f65143c4c5fe94cec148","metricsSha256":"d14f8c928b11877a789e940eb3c80eb67719103194210d030b55b06699034ed8","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2479312","recipe_identity":"ff926f67516efd77af709427c6cfbb873551e0254f0318aead147f87b4604ae4","loss":3.6730007590378038,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"5e1c4284f5266b7777623f6afef7ed0c0b922f6ad6e10c1077987c40b0b46af7","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"17b911b9a3bcb229e530a57f694512f808bfc5bcf9a0ad0199eabc2b17d44aba","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"5f62f6658bbba7654e8ae8db6d4238336627f9bcc7276283c13b387e12ff1787","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"14e0cb33060dd1aeb7436764cdf9e82651072ec78595ab1ddf409a2c978b59a8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"629b0674ae569a46e8a6b000b3da2ce4dd1c5f9962454af15bff131bb2852f81","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"99a54e97ab3a53dbad8e164722c2f12cef084a3c74a15d600a23cfcbb46eeec8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"b1fcb4f5d87184107ac5ec30959abb5b2ab3d9ca95077d1f90e57ffa82190715","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"c20eb9279838ca32853be6699d1e1e80901c9269d08957bc1f36c080186d995f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"a076b4d08093ee5bcc72719a97ae63825d1667f1202c5247634683a1dbc91e55","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"1cbefba5bbaa8208e857d933396d3850b43b19d60cbf6ae200279be75c44b055","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"a74c7e8d861cfd44c7a7383586bc77a34ffc8a3880cac75f39c9715f43c44756","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"cf76539b2bf8268a55cdd6826607005a2ab37722fe550c85f25e19a9705c0f6f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"f0f3b6b118d68b99088983c25b61d937da0e648caf26673f2e05d1c48b89b1e6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"4270f6224dbcf554bb0855e65c27023139d6868893260c39b17b2250319f16df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"404417e1f21cbeba92a9c80886588793470826230fa6ec7edd6a7100cc9269b5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"887f33aa9f095259cbb3f22db6f0b01f03fe7d6608d48baf79d04ed76e0785ae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"e1324246487e3d9874c620a590731285a9e8240eb1d4075a880edbae379be76b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"399be5551e2754ec7f20b194048299378c61f7c8558361e09772907454c2a59c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"ec9aa15220cc7af44dd44cd4ee5a57d65f0a02bdf3b35d2d251532407c63254e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"8b39fe3ceda0832bf738c69b611acc67a9d4c4a23916951d445ae9b429c52026","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment.json","sha256":"9f4da32d2e0f5964fe20f568ae284692d965329737885d360f4e59662650474a","bytes":5339}},"requestSha256":"3b2464846d5c9825fad4c1ff553a2d438d2c3d250c33d16d13f84a8d4fbe7a2c","metricsSha256":"bf01801963cd6306c81f893af9662e154fe12e0947e62c083daeab5188178dba","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2479313","recipe_identity":"399f27c41585d31356643a3ef5f9e5295438ce334a92f0a3c16c6c44a824d908","loss":3.6717062149905817,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"e694948b51b39f1d803a939981fcd016e790ea967904b1d4dd613a44d1b6a42b","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"068a274a599401ed905c9b4135c6d36592a893c883bc31eb054796dbbaf8ca77","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"f366193fe73fe91c6a5cc704e691f0b7e8da095855e704014938b246d7c9f101","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"d63c0de0ceaecce217a6e762533fbbacd0bda980972c4233332e3636d15f1c20","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"6c5d364e8606ba4ca7c283ae92122a9494cecae22a7eb0e20f56bb228e6d4da8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"c4f715d151be458cd363fe3a5c33d140a0b69ad01e453ea9f9a61e27c76f7293","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"fac3f52aa02d8b359514ccbb8d9610a41f68297013d4eed3dbc1621d3f4cb5a4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"3b531d9a706c75741a62dfcfef19ca0208c87960010bfeac8df8f8b242235b49","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"82737636ca448840fc0a5586d35b20a6a95d18aecd19ffe64fdf2e6799d1ebd6","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"2f3effa4ee2e1bcac8c0d6340d3600e87bb291ad19cb4e580dfe162fb7dd6427","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"6e5ed5a46d8d8833cd2a43087e512dd48f8a262bc06b31e341e54db522c95916","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"e9c90858d2186e7ef66af46aac2938e75847fd541faeb3d95fa12159179ae77b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"ea533bdc984e62b76ac96e22461e865c0c546b34b941f6d2fa6acef9f3671bf0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"0f11b3e419db26367561d8b10cbb4d5b18f293a315352281bd6a3324b8abe1ea","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"7bc1b425314b5d60b1b9cad6f59789f0cbfe86de1d1aed7faf732aa4539ed412","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"32ffb3f19367d55a0e5f3a23c2e0a1a7657a27a7f80391dedfb2ec67dc0ff9f0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"d31fc2338625ea62d25bfa005e0947c481ff3832cbe44daa21e40c00b4d761c6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"53e0567470ab77964f597b491e952563dfdabcda48bccada86ff9be219d2ca8b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"05f0b754996a4f5e927ca2e6f9d1e4a1830b96fc7dc36b3b51b31bacfe56dd64","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"1fa6a717dc70bb6474d70538905aa94a32fb99d11901997fd073232fa19646da","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment.json","sha256":"e53466996dfa598544e21aee57c0a8a46cf5b052f7a95506a15c2755cbfc7732","bytes":5251}},"requestSha256":"4c7ee3bf78cf7a425f2a0b30328420ffff72e8026b0231fbe3c147a740d452c4","metricsSha256":"24f2e8973ec488f89d4206c633f9ffc2e6f2662c3af80f7dc778c385b3b45763","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.01-s42","horizon":20,"attempt":1,"job_id":"2479314","recipe_identity":"68124984b08d6ca50fc86c13c1ef15e151088e83e1883c00d321498dd28d1b0f","loss":3.5950639300065372,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/result.json","sha256":"98e17424767fefd7f67c57cf33a09489c54b0eb2cf55c03fa07e2ae5034fd017","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/final.pt","sha256":"e7cea8c8d8b8c79f16d670e434d16ead38c25ee6479965cd72cb5b0cb2cd286f","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/final.safetensors","sha256":"574d8ef94797a16040478d1f1a881dccdb566d734b58d0ac8e8fa6df99cdf134","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-000/final.safetensors","sha256":"f61b410e80cb920dff32e31f5b755597366a575cb70ab584a21ea83284fe3219","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-001/final.safetensors","sha256":"98e518eedb142a6a929fb2f865e0725b14acdcbf1281328638080b006a5f17d6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-002/final.safetensors","sha256":"b90199fffafdc473fe411d40c571c8225ddbbc3cd7d1f40ca6e94a1e4f9f02bb","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-003/final.safetensors","sha256":"166a065257081f8de7b3da30d1a4c77adddd970d838894d823f0185d97e90f30","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-004/final.safetensors","sha256":"16c034f770ead6cd6e830c0a78a5db431a692ff36e54564d00f358d874c5a0c0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-005/final.safetensors","sha256":"dee12e1b64e08a6df7bf594dbc45735b938c1f1afd7d5e230ce76481b4d4b1e9","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-006/final.safetensors","sha256":"d36096f88beb2e4b9590b7c78be262aaa3b2b073bf0322717c4eef6cc3f316da","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-007/final.safetensors","sha256":"e6620167e12ee96d8dd65d1f72d346b6d5f220b3f4619917856d891cc4f7e819","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/epoch-035.pt","sha256":"fe2b812e7b04f07b601e455f1e52c56968bde609122e590150fcc22925cee09a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-000/epoch-035.pt","sha256":"dc0d03456ec6da4db9b9c16412deb8be3f2ed73b8dc4ee2ad52b2b1cbbcb08fe","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-001/epoch-035.pt","sha256":"5dc25c6e133c92705e13893cfc6c8453baf578c7d0f4d40ab51a19e98702c987","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-002/epoch-035.pt","sha256":"46dfe7c9b4b71ed056afcdbe67fd2ec04dd4b2941106f61b4755c6563e6a490f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-003/epoch-035.pt","sha256":"c811ba5365646f5eb79f45c4d21eee735a82691f0b40ec40b72d70b185ce977b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-004/epoch-035.pt","sha256":"b7077a66e77e10896fe8c7ae895cb9f5353de083462ca8c89b539c14dbe9ff1b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-005/epoch-035.pt","sha256":"96d4779b40e887f0fbda512e1906cdf695145c352ee1b81ccd77654d038fbdba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-006/epoch-035.pt","sha256":"ccad9290e7daf010f00e2501639e10c875d8813f936be0d7257cd834a9f984cf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/node-007/epoch-035.pt","sha256":"6ff1449c0c0fe022f362b5fdb7c91dd6a88e28b2c490aa55eb4d859484e1d8c9","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s42/attempt-001/environment.json","sha256":"021ce0e10511151914208c7e5508d5e659310d1644c56ad1c0902d25fe5b75ea","bytes":5308}},"requestSha256":"f42e2d43690fdc68bd8296fe9643855739a7970b19768296d7d7903b422da7db","metricsSha256":"0c25aa32c11fa3321aa6aa8e74a2ec192f8a70f0b85bb6ba73ab622ecca6a494","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.01-s43","horizon":20,"attempt":1,"job_id":"2479315","recipe_identity":"22ea9ee9d373b2a80040203397266fe8b4fcb032fbc1f977a1282e4196549c44","loss":3.5922687209068687,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/result.json","sha256":"72c723a32436e6e6b2aa6daf72d75ae73fbef95426df32bb6f2117ab516dfcd7","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/final.pt","sha256":"b673ef5a3e3ab2abd14e0a2a6c170f2fde6da5a0a4fe778beb558d9ab5dc57c8","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/final.safetensors","sha256":"a7dfe0c2f8d74e0e9391de530178d6ed95f62b003bffb1ec277793aa53b15a95","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-000/final.safetensors","sha256":"78e94f75358c1621285be3dc834c6c48f4566ebcfd20151d30ef4070911667db","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-001/final.safetensors","sha256":"c465e11d3b0802b17c723895a811f28a3da77ce91d020f2a0d3ed48724694a64","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-002/final.safetensors","sha256":"3e591510a60ab7d0f902418b69e0286b4e0f825f7450f6e8be74f85420a48352","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-003/final.safetensors","sha256":"c3bf9ff33cb7107413c8dadb395d3725570757c709a255726066e932ad4a2284","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-004/final.safetensors","sha256":"e211350d91b0404e346346d7d5a7f255e6af7cc79f0db0f2ac859e9dbcfcb011","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-005/final.safetensors","sha256":"92e6e94fea29d9a72243bf4636474b2be55ccb48dfe7fcba5146fbe47c38c0ec","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-006/final.safetensors","sha256":"2e5752c6b485ae7a7a632810ab52f04c3b8fdbddcb066a845737b79a52fe8581","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-007/final.safetensors","sha256":"92f98ccbd47c4f457ae14386ece8c149630b4de3df906531bbcbcb395865e086","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/epoch-035.pt","sha256":"3bec478cda686f24a1837e9cb0af97022c47310584031f7698cb04011190ef33","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-000/epoch-035.pt","sha256":"be5e897d05f07d89d96ed33fc47aa47e42e59f49eab4c4a320392255107fd827","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-001/epoch-035.pt","sha256":"1c0a372ecf5a2e084d72ce9e2c69b2e194e3888882a8146cf959683a75dc7916","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-002/epoch-035.pt","sha256":"803cfe139c0968e59fc87e87dfe1a172cdb88d7d67945248ccb715ad071c95db","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-003/epoch-035.pt","sha256":"a1a6c7a07a83196f13751d2307bf71c9bb18ceb842270d105e507d7ef06ce0ce","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-004/epoch-035.pt","sha256":"56f7692352feb04eea9e8b1dc33bc998562eb2c7615d964af32179fb941677e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-005/epoch-035.pt","sha256":"2e2019fa2c8c78af2f3429b8d51e6a435060b6e1d73b3b4d38c34890d540cd8e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-006/epoch-035.pt","sha256":"285f156af61639dd1efdb08ae2fee6c81eb592d06713934c2e3a1aa448f67484","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/node-007/epoch-035.pt","sha256":"cebbc943f297ea07f49a2755cb474b5591073b4a76f65cb09896bb3447f3820b","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s43/attempt-001/environment.json","sha256":"78cc065d0202bee290979a4ca05c841d4c81d8557bc06ee21096329e18565503","bytes":5253}},"requestSha256":"882e6e649e3e28a5e738b768fc80c16617df38bce306c5a6d8acabe1d0504eba","metricsSha256":"06f0037827070e0814389b93315f3e81f216e93eb241822969384f277236ca52","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.01--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.01-s44","horizon":20,"attempt":1,"job_id":"2479316","recipe_identity":"f608bbdf7016f91dadaefd58370addb32759cd65fae1cea2b7cf3b8a44e34821","loss":3.6077726903230727,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/result.json","sha256":"966487027e578ea96ca27ff8c80f8422704b08e04033aeef006020f17f066529","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/final.pt","sha256":"d93c6d2fbf988a03956ce745009b5f689651a1f02dc2746ce1f5944eb8324b68","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/final.safetensors","sha256":"4a037142fe08b733cae6f9136f7f983294f7f32179a3792f46d63616865dbd0f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-000/final.safetensors","sha256":"95a01bfe6e97d70e5968e037ff15c53adaf6dad1c30ea196d488fa20123b0155","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-001/final.safetensors","sha256":"8eb0f793887299f7e1578e4170ea767b271225efa0fe2e39db545c5fe9459434","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-002/final.safetensors","sha256":"04fd33dcb3ea166418b64fc7595103c000808d4c997e5e5183539bfef677841b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-003/final.safetensors","sha256":"c8e09c69ceedf40a22463e8f1e5006562ae64e5002996a237adf3fbf4c739cfa","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-004/final.safetensors","sha256":"d11e8b66c5bf58f385e170bd79be2537ffb176a176dc33f0977a7904dc9222fc","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-005/final.safetensors","sha256":"2e67d9b8d543b7b8a876c0123b64075c1f54640919c6ff4457236e6e965af2ed","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-006/final.safetensors","sha256":"efe0a0e327fc66b10db240fde0d47a9d57900f7fa52cc584fc3ed43ce16c45fb","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-007/final.safetensors","sha256":"3998ceaad469e66a3d2c2427e105e41ac13aeb4bcdeaa805a3df8488e60bab3b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/epoch-035.pt","sha256":"ddc38a686c08e1fb87ad6fefe653448e1a22542e0d8205ca62b89f1910c1f992","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-000/epoch-035.pt","sha256":"84536f6a639866aff0b4a54673166484779f25317dfb27ff5f1328ed642409cf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-001/epoch-035.pt","sha256":"6a5fb72d849ae63bb1ef65958c05b4d0bd6999f7e4a859bbbeb0eca2fe15bdf6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-002/epoch-035.pt","sha256":"2da41d865bc19f056ac2e0f6619f2e1a76bedcbf97c301fa5c2df2167c98ec74","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-003/epoch-035.pt","sha256":"db879f1f2f5cd7b80422105aefbf137d8c8cb3231ecc4bb182907e79dd31917b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-004/epoch-035.pt","sha256":"dbd7827b299b9cdce36afde9a0f34148bf899aba053e05821f27a182d7444b82","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-005/epoch-035.pt","sha256":"af77161afdbdf06ca0ae4ddcee994f2ec31b5db4ace62ac2f5d7f7192f646f01","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-006/epoch-035.pt","sha256":"5d13f3db7c91b1db47853ead4e1f27d29456e117dea6e23254a9651fbf0c49c4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/node-007/epoch-035.pt","sha256":"7617b48b59293a557fdc8bdae8b3ef8feee238fae1fd0473a10fb7bb0000c855","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.01-s44/attempt-001/environment.json","sha256":"b504965784eb000e5d9e00d5a151ea238161f1d4c5e9db7fea17dae0e18d330e","bytes":5255}},"requestSha256":"d17be515a9ab7ade15ace590bb6100b1038d16066bd2e0e2cc4390ceeb3418e3","metricsSha256":"cc0f410087558720be3fc9cbf89dc79f7f4065de2c3d2935ecc0f283cf7d32c4","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.008-s42","horizon":20,"attempt":1,"job_id":"2479317","recipe_identity":"83ca7d76bf5df5896c395069bc6116a79cc3f36afc5dbf132298ac22eb56860a","loss":3.602557061427824,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/result.json","sha256":"fb7d50d88143435906dd333849ec6c4ad65bd9636da6273b9b5a265e947bb2d2","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/final.pt","sha256":"eb9c508b5ef2667eeb70ae63d88cb146284e12ca0bde449ec664ff2150c76ad9","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/final.safetensors","sha256":"4173ef20ba0e75f69fc7477ff203bcf5e12f6d5be8ce6c4ad0307e5e1b230f2b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-000/final.safetensors","sha256":"f6eddb08b6ef46da7ad5735c21a9cfdbabd028ebb155894c035b39706b2c91e0","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-001/final.safetensors","sha256":"36338af36bd7524d17ca6e7f2070e6e59da4ab58ee591d46bd05f6e336c89c7d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-002/final.safetensors","sha256":"2b23c8cdb6f2c7d90115305f984efca19e261f94d37f22ab2f92724c7ea864fa","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-003/final.safetensors","sha256":"ada2e6d1ffedafdd8b9e606c5e166ab1f0ec1ca3b631f7a5f864b44a40489445","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-004/final.safetensors","sha256":"072b40e300ae4bc54061c4beff7107904d06268bf0bb93714ea73dd2bf3aaead","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-005/final.safetensors","sha256":"b5ea944ae5010f0f54f307c4feb26215d63fdf464660298114dac5646f48f679","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-006/final.safetensors","sha256":"7fb62fc8c517d1c7c2992aa2e02f8c6232c0ac338efe78021c63c0a8ba648b2a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-007/final.safetensors","sha256":"f70fc9d188f44791f84e2675a03699ad077c63f15fc39fd2686d0fff36a92bb8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/epoch-035.pt","sha256":"ec730998ab7e0038537bd951f4e81085af463822939b7319318da86af4dc0dd5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-000/epoch-035.pt","sha256":"421af11b4a8dc10a6ef03d296160ce634801cb27f316e9ba1443b5eb9196b9e6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-001/epoch-035.pt","sha256":"5a9938578e6598aea648931a74199a7ca1f2d995c8d154d1e8b8df80999d09d6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-002/epoch-035.pt","sha256":"8431f32e3c67b8cb57bb86ebab2809b93f65327777868e43127ebea19df84c94","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-003/epoch-035.pt","sha256":"e4684dee10f4a299d20cd12bb4fe5355de041259bb8c21e583a695c6efb78214","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-004/epoch-035.pt","sha256":"00c0aafd9a6acab7359da4b588c7cb7ab086cd80bdeac8e4483c8241a215f124","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-005/epoch-035.pt","sha256":"5eab9c4faca41f1081d7ad421ecf08f8fa8e4d23976bd5834c484d64390768c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-006/epoch-035.pt","sha256":"2be3e024e622c87b41599434e471a3d0aeb613588af561b24fe2c7be9def2077","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/node-007/epoch-035.pt","sha256":"d22bc86a3f9ed93452324bdcd4137168c2803ed79ae8a3a1158af75c8ea91ab0","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s42/attempt-001/environment.json","sha256":"3356ed77f53bce173d7f7cc17dba26758f251b63195ec73c7472aab90877a1bc","bytes":5341}},"requestSha256":"08ca715f0ad4de321dc278d56994c40317f9a7408babd91b93214dd46b0426de","metricsSha256":"91d3e9cfe0f362edf7aa8171fecbc47d5f7034e003ee487ac6cff2f3e65b25df","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.008-s43","horizon":20,"attempt":1,"job_id":"2479318","recipe_identity":"1701e84c843595e224da362d0e354ad6ecf4762680f02f8eb85af7b457203e87","loss":3.589014955110382,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/result.json","sha256":"01ecb96e97c298c5fbf83702af051c926a79963be797851e6961f474662d94c5","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/final.pt","sha256":"bb1ba12f907234ec10cb1e9c7975d89576993793be6df4d06ab27dd72802790c","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/final.safetensors","sha256":"be13ce273d6a555055aa7bc49c05884a7b0556959cbdf587ec31175d17804a64","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-000/final.safetensors","sha256":"794a4965a5afa682afa5f4112e40c252e83dcc12cd59eddc817567241521de55","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-001/final.safetensors","sha256":"676888ab8834750d0ddf9b762020302e641d038d1e467fe848d0f87545428687","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-002/final.safetensors","sha256":"2e01cf8a51c4f539511181e93616d81a3a137044484faac56da4018348821132","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-003/final.safetensors","sha256":"7b1f429daf3de4df8163aea91c898a6ecfd626e68f3e2d93685d6e29f077aca3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-004/final.safetensors","sha256":"f27a2bc229e0227eb69c8465034cb247030a19b18cd44dd6ca1e817fe2f9399c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-005/final.safetensors","sha256":"93b18f0dcc92fee61defa3ebc2a959298851e83cec8425597ac2111b221a3165","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-006/final.safetensors","sha256":"41dd17beb437f093bf18d8e2e405349120b579b20e22333f884889e92a9765a3","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-007/final.safetensors","sha256":"74f4e0f87b20dae2ceb6735a8b71b5aeab2e55abd9ac1879acb9c75784349c13","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/epoch-035.pt","sha256":"a25c49142799b55d14c2c873a534d1af5a5f41eb0313d1f1db32be97673cabb7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-000/epoch-035.pt","sha256":"7a07d35d784c6aa557466e385d85e723eb203aa4e74fd72197bf427df07375f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-001/epoch-035.pt","sha256":"ec30dd06bcfc8aa252bbcb5f28d9dcf6f8e786b51b8b624bd4c78571e6ee541a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-002/epoch-035.pt","sha256":"419eb9f35bd78a765396803f3a3ecdd1bb39f0c822a688d2a9f15e5914736df7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-003/epoch-035.pt","sha256":"577c4481e968794ee69175064e7edc6ce4d0338f828e738c93ea3f2a4f6f946a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-004/epoch-035.pt","sha256":"35015815511597651884baec346a5104f9d458b013753da22fb91c88f5cc964b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-005/epoch-035.pt","sha256":"6785623a4bf14d909d35cc6ce46f008503d066aefbe7244399040d0112f8c91d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-006/epoch-035.pt","sha256":"ca4b4400d249ec2761bc4ba866bdec7bce0a2201dea4cf3f1877eaaa212e9348","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/node-007/epoch-035.pt","sha256":"e243710484f1fd3d41d9afbea3acbf50467dfaa8daa37a155c14ad21d811a141","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s43/attempt-001/environment.json","sha256":"5126574c87206f44b970e5eb6959dd4446ceff55644a12c116c6cd20d2276708","bytes":5341}},"requestSha256":"b802e6d5f82f563fac37190dbccb49120e3ac53cbd52b90facd91eea5a614bdd","metricsSha256":"287678a2e94c743da3ecc6522de2c0c4fc438bebae4430763e88506e5a05f6bc","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.008--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.008-s44","horizon":20,"attempt":1,"job_id":"2479319","recipe_identity":"d3f1de2b7185223efbd27694347e4f488dd4b5e13e97f0b4e62ff7b4393cb8d6","loss":3.5940841008689515,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/result.json","sha256":"2d4517b858c7d4c65b0f1d16028ba41e48ab2875ab9826f647661438a61cf620","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/final.pt","sha256":"905fba1c200cd52f0b78cb5f81e98d9467dbadb6223a1e72029a3918da929ea6","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/final.safetensors","sha256":"f24ddfdd2bbe607f3df2635bc7400b8487c1d96f8d5009285a877d39af92f0bd","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-000/final.safetensors","sha256":"7284cbd8396492a8b284f954fd289ca402998397855b26fd8dd172d99b757b1e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-001/final.safetensors","sha256":"c6cdb2dcfd664095cdf0206080558cbd6539d12e67d8fd568c1b248a3d9ae6dd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-002/final.safetensors","sha256":"d79974b21acceed2b9c63bfad72b2f68e46f0292ed538fddf08cbbde9390eb08","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-003/final.safetensors","sha256":"73b636b7b4e5b1a5d952e8b59ec1700e380c7dc77d64cd92cf8b30612a31edd2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-004/final.safetensors","sha256":"40e3e3b351e6aab4a7a3eb23f941ee0dd83aa44f0e1b3e54ddad04912f303ece","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-005/final.safetensors","sha256":"3099b877be8225dddba775fdc51a02f932d8390e4b390b8de797292ab0a4ce4f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-006/final.safetensors","sha256":"3a8ab3b5d2bb8964bed42b937e680cf6d824f7806d5aed94b5aab88fa89152d8","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-007/final.safetensors","sha256":"a4bb9493237d0926d7c0824cc72378dcba4e3e13428c98c4325a8ce758f8b237","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/epoch-035.pt","sha256":"5276aa94fcae8703a503988f81ff216bae034434f1c256b0a847042ded9d94c0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-000/epoch-035.pt","sha256":"87e4a45ca798d5bff305cf81332cdb1f0e3d353e5ebc10f80069ccb7e500f5a8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-001/epoch-035.pt","sha256":"b5f502d5734a5edcbbfbd083e25fe5e3fbba299340969a7a2170261be728cb79","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-002/epoch-035.pt","sha256":"09347ff4458d0f0133c50d37ca3c016e99cc899ed658972937e6262d6b919ab2","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-003/epoch-035.pt","sha256":"3ab16dd2faa9e8cb79cb2d03dc3183eb3067738004c09966bcf792687e5f205b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-004/epoch-035.pt","sha256":"9b303e96390c52cc8641a96dae1826151e950bceb6dffc1269b19776ab88c1c5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-005/epoch-035.pt","sha256":"ae1af59d5a8fa073eaa8bfb299b532efecbb2d1de0df5d032435aa296a897285","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-006/epoch-035.pt","sha256":"f288a9951b7fdaecb86704616c84f6544e099a1c4d80108715fce66ffbfed27f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/node-007/epoch-035.pt","sha256":"d9d333bbf444cd9926ad5bc55c426076def7551b198aeef76fb65f4acb6b8bba","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.008-s44/attempt-001/environment.json","sha256":"90a9f697e7aef7f463f20efd297bc8d1408d839710f825efb5dbdb2b5a1a1da9","bytes":5255}},"requestSha256":"a92d72bf5d6aa60d808e5bf1f2e35cff347f1fcfac523adcd265642a5a69436e","metricsSha256":"ae36f37eb1d6dd61659fccfa32f7a3bcaf269a9a372c7bada0fe07a3d21fb054","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s42","horizon":20,"attempt":1,"job_id":"2479658","recipe_identity":"96123d836996f5486880cee3654a7e6105fc12ee498db3b6b4cb6198ba32fbb2","loss":3.5881057930940465,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"765c42f27df511ce17d9b5fb84d3daa01231976a16d9266c3b9e864f15837eff","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"77b2b15bed9aaba63d7a808ccc563bf3b7984ac4595978169981ab95c3ccee25","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"6e130aeed4f8c8be3ae8ec37fb95210ca82ec115b3ba55e9e6850c6254c26b41","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"64b52f958ffc31d696cc4addc88dbfa5c463b61524fea70b01c20d14fcb79728","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"8c47d6b878b80177d21c52a6e8746d647f10727bb7d78e31aff9ef8e0526c7c8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"a2e1e9773b39b214251eee88ed4f64e2df3f4fabb1f49ca8cbb9fc1cd1adfb40","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"69c1336f792588b4ac8a9d7ad2569426c697fb8f6dd72c0441c5345637eb0f27","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"55f97faf7912e3fb07824507f3c1ece1983ec2149687e65f719019d1e0ab7c90","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"d972dc62ba835219ced884000049cbf497582d2257ceefec6fc79d66c1838d21","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"8c640a5c6d8b83c940d3f8b51e16902b301cea7092ec3f2f55904bf1cdfb6203","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"d52cd89e8eb10f53d046b3420f661abb2ad3cf918bda8832462a54f5c690c95d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"66c532162195e9718a70392b09bf8d1dde4ee42ff114d09b8149f942d0933c2e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"7b123f5217dc537c2e6a290633784a82fa929c7b61a47f9b88b4545eb079974e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"09c632516286a0e6fbc106b7911618f1f203f157a66ef0847ddc415b13ece91c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"dbfa56bd13a3d4f93dcf4e80629424a3544442e4a3a511d45403f0465215dd66","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"fcc618f6e2b0b5e03dc9ad9980fc2de4497116d4e121c4672a2435cdd4d0f300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"435acec51bc99985f089bf0b746b550672f8a4a20997935c1f4340846871936d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"9b5e58662d265f245671e5e7e9a74320b7a65b6a71018180f1a80689f598e2f3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"6fb2a1c7907061aebe9a6d19fa65adcfa4ce1b51376151e57f618f99ebd21383","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"40846ed038126f0293518d4ca276639692e51744b3bd176bed695efa05d923ec","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/environment.json","sha256":"aab8046113790411bb525b96be00d5355a00e918a939b4edffb23cf97b87949c","bytes":5311}},"requestSha256":"1f78eb7a8116a64e43c5a8833e55192a2b834e8c63dd37e057eeb4eb6ed2149b","metricsSha256":"38469d08f80a241c1df86f20e8645ffc8c9d5367149bf946323ca7b5d08bc932","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s43","horizon":20,"attempt":1,"job_id":"2479659","recipe_identity":"17b9da036f44b92b3f9b774b83f43a84cbd323aedac1ef1f2469fca87fa3cd25","loss":3.595977846361822,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"eada167baca39ccb2035df12e84712642fcc1401bc80cc0c69ba60ab860cb6db","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"5306692e02fb4bd09ad468a4f0444a8d523c076b1edc48f9dc2c38aa0322f0c9","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"936b01d9ac706a8188d2d16f3537a6dbb7369ec3a41be72bab71a65d024a0f9c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"14337cf115ddab4cf4cb2e216733b21f7601882f09e06b0422fd841cab0c7da7","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"01f6b602ba366a82c53d41749dd1046b9491cab5334664897a7b7f03b8577169","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"d6ada747fac0a994a3126cd48395408278c6a26fa87ba9e8bfa5a858d94d0673","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"1b51d7970a2b04d26be4e65df9593e05aa0fcb9a018ee19e1626c1cbb2049118","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"843dd1fb40033f42f00a8f4a687873947192259c3c3bb20b0640e8d62e0cba29","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"8e3a90613fb94a29eb35f23cc7a3430eec583d29f0d43a7abc7a1c5aee725d68","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"0bf506b3e63b98547330bc045aa3484b5e2b28f4ece28753cc4963f7b0a3ffc9","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"d117cb2688dcf4885ee4ef06318b9a5ad76a93ee1b5b749ddc640a63f59bbc32","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"580a3c0cfd5cbbeb136c9cbdc582a4e4d5d172b9ef025b69f3c184f6a267c86e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"ac8e1eee6a6220f1443e0d8ffd4951b44afaea2d2910a48741fbd0b43903a630","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"972037068a1ceb7db79c1c1df82624ca54bc296ff188bb496f55645e214dbb3f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"f7807bab04680a4550e024a5caaba297e1b90b27ec57c563f63c11b0b585f781","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"b0b2e74c35c5052738b376d9b8a1f71ec7ec6f4493996dfded75ada3a0b5000b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"832a6d05a6a9ef0b7567e6e93659f2dde8cd9e1e1e89bfd6686a0c3a26ee1d80","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"b65f72a8841a446f6c2d5d701cc2dbbd75b4cef03b7e10a4d5fda22ae5698c43","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"ac2316e94075e7475890eb0f863e9a8d5c096f6bdec86bc3097c51adbb4ba9f2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"37633ae42a558955bc2086a08b56ae2699b21abad4ed91675a0759f146e37351","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/environment.json","sha256":"7bd88c3c7326c28518492de829809d1b3431aa8a684bc526d067fa741655b9ad","bytes":5248}},"requestSha256":"5e80c8c5a2d9baa44de7b2550879f1f49b10c39e1fe1c9f348c56cf22b9afa74","metricsSha256":"933a9d4ec608d20c4dfa86dcd45c86504ae259985e17f50973fb4d6a6ed7599d","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.006--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s44","horizon":20,"attempt":1,"job_id":"2479660","recipe_identity":"80800b13f26159de620f6538a469f84add13c4ed58f965ea6e3fdc61018d57b8","loss":3.5953002432009535,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"d0ae85842dce3df5d147c75215fe4ec3978a0f06e5b3fe444248d6d1c2ac84ae","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"0548685182015486a73412381c2f9208277d7d598dded30f13bbb916af714a95","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"0f9d85818422d830226d0d75db129742e036b9ead27a3cbb2f3b951b9aa47b11","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"0c935ac8c640e1a6dee8183b3147665fce43bf8636c8cf62709b5a3d49ca63b5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"51d1fb61183c492c1856937dfff7f6234d2f19c95ad763e8ebfc245f7ea38224","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"3117bbb758088e81d81c026b0c81ce8c3c7c5144c0c7f1a46af3c552ba97613d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"abdb7f15891030af77240d371e96a4826bf48f05fcdd7d32daa06bc812e56f6c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"26cac6f6797edc2e44ea36116da3586661b531e2fe3e876b66eed02da396ce5c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"9666ed2653567fe26378bc4f1463f98c96f693610f878fc069fd6579075818da","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"d32c75486e7a16dc54b97fb1f6ed629c231c5fdfbaa455c28e7bef73583312c3","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"7046cd6e7a1dd7979b55a34b40fb89550d11e7e69e3331ed3a4acc8c0d960381","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"4bdc97f8d4e80b5d4e6e5a954cbed355e145b1adacf7c4b8823b77bcd40aa6b1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"6cf894f9a812e34dfd74c56910b8bfae66007ba29e779469c661197e4f82133a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"7affed6c7a9cb0792f0c729b826a1d6d865f89e428baafdb66bc66156c913d4d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"5db51acbc00839c2042dadd31eb3245232c052383f8a365eadcbc68304fe07c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"c08027148e461a1828536561324edc0b55b42ebf6171981eac47887f697a0c17","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"cda178d4824acc44a51599077ed20ed3ca47bf935d88afb85a05be5654925e2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"04d3fd6e887404e5225ab74afc30e5710b98a634cb471d8118805ad2f3b5d453","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"5956cf06a82e1b3454f606bfeb51d21d55b0e6eb0570b44e694b1fc92a07a771","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"310da54d816e25cb710a4379395e2b1d3b5d4ddff297e3f1cddc305856d578f0","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/environment.json","sha256":"1dbb88c3e0fcab17c99c0c2d09f047d311c5cdc2d7c9b0c4c0466c6814c9caf3","bytes":5338}},"requestSha256":"712c9743559aace44e3f06910565e5dc2088b27df17e68d347d6d30fd19b3cd0","metricsSha256":"614a3abee392bd26f6bf28d34193a30cda7d316af79dab99f55b0bc64c7d5260","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s42","horizon":20,"attempt":1,"job_id":"2479661","recipe_identity":"efbc0fbecc6cecd34089980b75314ce6b374da0ae3e094366c22a621bcb9eb96","loss":3.5957114146924805,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"64b571bc1bff433300d2490aa5cd8ba0311b825974f939574a6ee5cd4c188837","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"23b6839e03d16a004d5ef9e50bb97cb94d0cedfad74d01fac88f5ce67ddca336","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"ddf06107a46068ba444938dbf3b79bc1e7bd266a8c8fa14a623f504fc47237b1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"503b3866d64de13a1988d9cc2fc654d6b50623b297f80231dbc22bdb804663e1","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"eb587f9a1f38e527df700db30ee367ce9e1bf31e0b268b0dfd70301e4bca4a77","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"a81cd29e13a093727e2629abd0f1eb7ada158f48870ce9d090f297802e339d4c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"5c73efbfc74ee2d50ba9f31ef3b24bd7a3efd827fd3698f43c15a44633c322e5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"3ac280cbcf9ae02a8af55c85893f5b5c843dbd5f10862962c5e0176325a93ea1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"9e8c627becb02a74de07a55d2da533e40c8ea43b84b7b7f8fb514bf5637c6cbc","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"38ea5cc5f436b493cc9629df25aaaa4ab17f33818cbd60511bb5d80d0c7aa0a2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"776a1f02aab93dcfcdfc5a1b027786eb7ff17728fd4c09c40d12812835d20de9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"c3c0f3c543db4b8c6679837515c9028cc66e709a2e66bf097502c79660f85755","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"890f237d62e89fdf4ed0c320db8da076f713dcf086545f372846e08f9661a788","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"eac29f4280c08e41b08d5244f31c0be38573b0e1d4f151f36a558168f5627369","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"4b5df2dd18f53df6cf6a35467f24364b2ea597ae6a2e0517598190ad4efa48e3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"ab001b3d21210047e151947cbe3aad140e0b2318535b97ea05e4557d679fa195","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"fc7ba447bb5edd7f76d670055901afb4a628bb3eda3723a33401f5d3804be15c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"12eb7204c2912437adc61c876fd724ab152d1ba94b907dcc5324f8b9755ce726","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"1826fbc18fa62210bfdb8b15a583f76a6151995114768658e4fa39f1d6fbd7bf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"fd01e0ab0b85db461935a3618e9b459ee7a26eb57cb5cf7bccb187eb83091761","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/environment.json","sha256":"c007edbd978da2e83a629bc0737d906d78f188d88f2e7edf4854e6f2b458ab85","bytes":5255}},"requestSha256":"cb882397eca61ed3285c673a35d2d0601d9ae2b20888cfe356aacc21d18712fd","metricsSha256":"d466f61b7704068b4c731aeaee6b56e0877010889447f5be2e9537287c71c6ae","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s43","horizon":20,"attempt":1,"job_id":"2479662","recipe_identity":"c40cb7a9f9ba296534c588c9cf0845876f79e406bb939c8f930d580c1452a7e9","loss":3.5953545512546574,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"f87a24eacc765a45deeec52b33fda8c3bbd9c5aa6024b33d37d8595444405280","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"380febd044c8eaef02d9152912e9225ed709526e262186e247ec9fb6b220ba05","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"058849f314f46f4761e74a6c3b8f029225f4ecd78cc0c296f078935de1167bda","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"d539d3b27fcae700a72849036f893f6a77379b9abe1c0aa2069637bd05d3b602","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"f919ef609112ba7927f54dcabf94e29c903cda83ef771d887cb0a54592017ff5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"bd7cacf1d16e4deb58d70c13e453d651b1baa942338f0e22628ee8e3e805dc90","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"74ac2b8da6bdf98518350e9b07e994a506742cc0f4fbab86aec93a7b58b06f82","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"f943e0d156d7693fe4c261145e0967246228678004a4e3bf4c62d4c813b32d7f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"170988a36081a80b82e7bc9e59c4fdc9eb71dd8374388593d439ff0f01b73f77","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"775bec932dabed392ded931dbb258dc2ea1089a315b03772b73c1ba9949a6d4d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"0a58515a605639cdfae24d3540627a05c9df4061e475153a1fdc9ba6ed93aa50","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"8319f858936c41a9dfe28993ddbe3f785a26805c149cbab993e9e8f8421b2bcb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"33280b3d5940653d96f3c7fc0e1b43d158cd92294d40ba48d3f4bbe0f5fa7243","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"e896214c284b7058b9d067251b9c8f24b576394c33a9dfcd3b773f53cc12f75d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"05913ff574d25b08246d90a08f03b256ebefa3c2745b2929b0c36920a8b7b6bb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"8ddb4fb48ee91b94b240cab69c09bf71ab9843c02a5a8076c48741150fb77776","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"188a22a4a1437b9b29210bb1b1300ab107784da94ee1e47515d76c4a62072b6a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"38abd985381f12363149ceda401030c893db0d42015ec4add599338b21852216","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"98c17bff725baf9e7d08cb9124662b2ecf653286151013cfee47625236039850","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"cc28fe00800def3ce4979fe840f2e316b7e1c994bfd96ac9e197b8cd0e7829be","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/environment.json","sha256":"4afde5406e91d6af5dacfbb0c4b615efb91a6e62bee44eb249d1636e96306a27","bytes":5255}},"requestSha256":"3a16710230f9291cb18abe4d9c57ba50608c2431be75aa244734b2f635ffc561","metricsSha256":"2e1c8afd7f71e72f506561113a26a74101469e4a658eb10ec3980f0d58a4303f","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.004--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s44","horizon":20,"attempt":1,"job_id":"2479663","recipe_identity":"c060d02fb5238057d2ab4eec69a11852b15b5fe6e2de99f36ea8c876fe5f104c","loss":3.603553165567626,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"8be478316d50fe4206f26428d18d93af35c51c9447b26b859be1512da1184296","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"3a43bf510aba61f552f4c48eb5edb3f615bf3a8ab0da8abb4d2ed89ad9fd18ae","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"53fbf4134fd23b34804b9506ff8c78dea77f6c028ea87d758bc2f9710676f0eb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"282f96fa1f091ef2a462bf0342a47910ed5dc8c951095d4e665b325a3d33b2e3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"feba1b6ea03a10ba1eac8791f48cb474e8433ed2b877affe1ca211aa08906044","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"2f4a2abe6eb2a0d89c0e52b3bcfdb33f596944ebda31075cfe41cfcc0ddc2f7c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"0717541e8add5b4106065675f382b1c91d8e96169f493cfe14642380c98962be","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"170288c8c302efbbf3dc89195fd9f10499266109a528a34032e6d53466e03c55","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"1252df48b4445a30676fa14968334166c53df2c157c1f9b5af3a3b180b410469","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"453b6642244d83f101978eb2e64018d31c1edddb7ac1deb036fe9cdc68539fe6","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"ac0b54cd41669b2721de69c6e7f93b31e09be630e42261952d3e2f0afd465807","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"d06626565c613890b416479d9dfde00a07e671ecab5fea897679964b2ebd5b1d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"b749cfe0ce86229dd59d2a65966af74aae873e116d40cf520f2991aa29ef8f3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"8063f1f3483743e474659f344a4fa7807e4bfca34fcca0f722b6f783ffee976d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"9e39cde1a37bc83629c196db9d5746517f6a69d43a85415e5f552793abaa3b45","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"a6338d283902b413e87f6b570c7081f3a84507ae9240c4451b1194451542ac40","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc882d342bba603fc791d0a283497e1ff205153ad4564b266d78f05e27e50808","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"c3b0ab747902f16a150a3ec919b9fbac9a009bd97d2c45850a647602add9faed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"2f8c8b394d107169374d621ac9a48928cd2edc085317d8995fd0e50853c16247","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"3edf73d80b8ac4d926c1bd2d27f19c7a70d3bbbe2e2bf0051b8776d597c6c631","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/environment.json","sha256":"325a1520b7f20004e0b2ae95bb8543f76c3d2057f48b93752cbb540867b15878","bytes":5312}},"requestSha256":"beef6cd8671e10fa048ecc0dda018346fab70fb28cf5772644d6b65df8df96a3","metricsSha256":"e19f2db20102c0c57d711702b2a150273edf31579a3fc1c6d069ae455f1bbc0c","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":20,"attempt":1,"job_id":"2479664","recipe_identity":"c2692992567ea6b3fa916bc28c202b0d89d28e3579cc707c1695d6b5260a3a30","loss":3.5982387195327967,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"6ebb1dd69c9d60590b0efe955d9ff61f35ee1572e94c74b562015904453037e2","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"f9ef968d0b63aa814d7e52998051db7ef136b62818877ad69070463bab7a0873","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"7805ef31f51b4b2300f54ed2ab992208a948d7699e32546c23cc97274df62070","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"f38ec0ae2cf36816f6cf203000e2f56351e30b3c72967bac8470e2e879c69665","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"29395cbed5a3b5f93485f67e2c9db2a33ca14e154498ba78e5c1a9119df46742","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"c58257641bb0c149a743655bde2d49281e1a6cf2b4eef94d133fbfe7603ece4a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"f683038214180832c186f6dc543f89246de0d2d52d164b75ec86a0c53530e48c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"25a547441a2caf9ee4098d989f199817a0918486dc01bab1de918958e4ef2db3","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"e9d04d9e7248850fa89d4ca7cd7cfffa4771b7e7598b6dd3df4ce02cf3889711","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"e0a95e5b95df5cd3b7cecb91abc864ad54003b3f9f506d542b8d4e58d20cc240","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"69cd371f5c4e4ef3397e2674f6407bf2f5f2c06a4a1b27fcde1744ca56eb0c19","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"f73a5f2e2144ce0daf03ab63017ab1d0a8b13c54e710bb6df7980f1d66c28845","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"bd2dbfd04c4b333df0ac0134340d3d1685a72cc48c5e3ffca081c74dff1147df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"64db6878edb2692c4ab80c4e94aed665eba142c19945979d809624bcbcfa3a6a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"9d77ff8992630ae47038d52493a4a0daa35d329067430dcb112afa233a29e86d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"1c1e36ee92dda312cb8d21a3009a0f6b33ea8eca4549fbb9d695a58d8c314d72","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"4a2c295939088317dbb174026b89aa70ed7985bad37fbdc20a3ba7a5b12c8564","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"d79a4ef2afc5f2d4d357ba540f00aab845d69505ea2229a727f9d1ba02fd69b6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"6c74bf3be268f1c170b038ff60012c4e29bd94a531483d2649d50601bbf5fd09","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"7e84212c0f453b02e8bdabb235224e5cb21da6aa47bb6bae4d8c1eec2cb6c42c","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment.json","sha256":"3071135e36c569d140ee0bac002ef5ec0c081f34966c9b37a5085d88381edaca","bytes":5307}},"requestSha256":"0f35f0e91140e26bda79f6e4b57b80acb5dc0b66c87c3cdb50193ada1021234c","metricsSha256":"17d470755da372465dd0d71319a2f643157150d779aeb8f82928aa0a6160a724","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":20,"attempt":1,"job_id":"2479665","recipe_identity":"387cfe8165347ac163a74a0316325cee9154bebdc8446e2f7125f5e0da24ba97","loss":3.613193303011285,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"a001f4222310001e3cdc9618d9796a580a22dee40123f53bd93957688b88d463","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"717fb87be23ae7e0ff7b86a0599c407ad9d71e4ec234c6a82c57223474c0e1a7","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"7e9a1d34cc57cd7f23c27b1a8baf995d81e04fb4d5d0befc3ecea996ba8541a0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"cece43f21bb83d21fab333b472b14b9971663fe11fb870d2813acba2d1feafc4","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"f7c76064bbea1746e0af248b4bd16b54b1cdada5b41110a2ad6d05b0cddab4df","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"cedaeb6dee2a5a1845eb869508242d79bd0f10960271ffc05342c2382a0b42c8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"f64474476467888467ccfcc275a5251b2388b0ad4459a2814a22bcb71b2c2a2d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"745c7c8a017011d4e0a4c8a98eecc2202193160a17b977988ca6080e9baf068f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"167c9ec66805cc70a43301f8b77bdbd58f215a69ec128ba4f9929310a74a471e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"43fb7b375af3d0d23fa457b619dca8a65042714f3fa619a32cfef22d61a85bb4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"9f72733c84620d9e5c307a7986827e4361129a27414c13f0e6356b49e06d79ab","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"129d2ceb02f75a495ec1303492c549ea4427760a30353e6c1a838ee90203b0fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"650690386e6081a6edeb3ab8814be8b3e978e78697fba580b58155fab956864b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"68b9286b25ab838205587b2162aee2189880b17f0ec6d52f03f8c19bf30eac87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"733bbeea6cf124031f930c4e2dfffa212154c238eb987806cb0ac81a18e7e56e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"796c17e93a537b9ab75525278fa2204357dd344a547851a9b7ba700d6c299f09","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"768bea380d3c7eac92f816bbb09e332f2e175d8f51abaa4df57f87db8cdfcd56","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"9890ef791815e67e274645257c7c130533939ac8d8e8a927d925c3e13027320c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"65397ee439f3603e5e2cbe54ad35be996f991ccdd291ea17a2419d99a2742af2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"62769910541261c88156abc70005e30d1c84c02f118864b27bcbf10df0f60dc1","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment.json","sha256":"c6df8b4ad367a7a453cdca79fcff47400aec981137c1d9a2b0450a2ed42bd12a","bytes":5249}},"requestSha256":"599416f0aa2e64b445a1274e294a6fc9b73e6ecef0394ab0cb2131d5ca9b160f","metricsSha256":"e20117548b75f19e75a8c4d223232322547d7b0bbd86934d5ca336153463a380","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":20,"attempt":1,"job_id":"2479666","recipe_identity":"dba98e8ef258152b440ef9af80a2e12411c8eaeaff62703246ec15bbac7e2f15","loss":3.608397527065818,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"1f94fce8166a0d8d842649e5f67eb2909f185118845b955ac92b86bd936553a9","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"d70365fa6983bd4723df743292f771538547175d5a62e938b4f0ec4d90822339","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"d60e888a96494e8617f2d3bbba0164c4e388d65136886a18277c244eebb220c7","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"8bb54efdb65305515fa0d097b791b49ee72a1e253bba025438d0dd864ac3ec3c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"3f29634136db71e78636d3ba17200d88c1cd8240e92962a0a706480d0a7f4c3e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"379e56abaadbd46076dbf7ecf9bf66342f442c608cc30742a539c5802055e59e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"f489272df23595297ab43489c8e72b75a0821bd04b7efce16b5991909a79d6e2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"249d36956809c065027a35f6d132206444467b3c34d3a2e4972b08b271579e07","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"633520c6083bc69fc24ea746819fa92e7b3aca00018a7dd571a475269b6f01ef","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"032041db81064639b522b33b25a3c1ca4714f2b51be7e6c3c8715b8cf46e37aa","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"35c9df90d119b319637a34b375a832d876c397e0ea546b65b1fb33fb8e6d202b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"87ae862144b80158547657bf3f2e51f719b081a71fe040f3f6bcf7f40f2799ea","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"1c63e408bf41af32284de7945ea776fadb72497830a356651d6f2e80bbb22c19","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"54b7153daf066fae12fc801004ad4746a95f45d8d06ce7e5d6e2a86e2fdf3ed1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"089be6fff7b4ff85b5a5d75ed1ce8bb85ada215ea7f01844bb7d560c7fc00143","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"52962afded4ca1a0a947beacfa9cba20e3d3263b47b3a4265abf8261d7dc6c9b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"20ea1fc12d452737059de0a5967c3c8d887e4f1294834fa6b8819cde6f56d563","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"7b039bf20a76248d100f6190e9d315f41c43303fa8dc95e69354eddd3ca953fa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a490cfde6a1f8ba1c7c8da4a5a970640d466c6c004422c8e392464384c6ac2bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"406bfc3340154c95b9cd832ace6577a8336a5e601fd5534b5700c722f6c3c7e4","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment.json","sha256":"f0bd132e361aecefac483075cc820b871b475f69a06811711817bd3c116e5d8a","bytes":5310}},"requestSha256":"0f43c435a13f3c005315dec063024db9b8a1ea489d2fd618e41eb09a13185cc9","metricsSha256":"a23ae06645c5fca4b52f0e8e1d9298662cfd2d2cc178d45a658a3f5de748a018","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":20,"attempt":1,"job_id":"2479667","recipe_identity":"f810df0244f56fd9aae0688279ff5949bd8ee7595b5e82a82ff4d739136dbe01","loss":3.649128402327742,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"b46acf88e1a9a8bcf93332d5e5aa3059e0207caa3ab6d07c306a5170d6101875","bytes":881},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"ead7f377f7d8e97edf7e4807b994c5721f9ed262a806bf7168b104b379db90df","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"8dbc5bcedf3e1150421608f7ac2e0ffe00c2bfa0bf41d8c4f5d5101c7c7dc259","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"82cf97d8bbb72417427712e65460438d783fdcf62a7e04156d8548dc0ecbe911","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"8f5a9ab465c1523437d76b6498b6eb2da8e16a25e2d48343308057e8172a6494","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"e0889a12da1c6f3b53199031277c8e6f964fbce2efbec2d5292bb18d3904dc0a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"075517b93391791ae630c89baf4915776bf1501699aa5b7f08c0b603e4d3d66f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"18d77055a0ecb43f2e5ac2ad1a35b35e11e1e99c652014f80a3b49b73993ae8f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"535b15a5a59f739452b10bc2ebe273b7e6836c5f9ffcc91f40bfcacbdf3f8266","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"2eba803b948258f515a2500b9e11ad5833e1044f3db51e7da00c51bfb7ac2c62","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"bce52952ede261d6fda3e5f60dd842f2578b5d8404739336c0e1b78c9ffc0aef","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"b530cd36f8de24ed50ab6206f85dfcd930ae1847b1e8da4050434857876cb085","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"a3aa32db6ec00877328db1a520010a34441c66fd9a88bd125a82e05acff1be4b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"a3dd1453259ae08b189ac0b135fcdf5ce6ea876dcae10a7f30f06b346310a69b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"2322ca0cca3b2b3c3171a6a988763125a2e2ea9f18137171efb54fc25f416daf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"30000f5926c32549bea14abfaea9028544e8b8ae3d7c80dd1419796c733150a7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"3549e57fc3ed8a5d831a6c9a791626a52b0f9f4b6143024de849820c4b323c0e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"5f2c1a0aa6289a7786f798cf55aae2d6a6c79b54724cb33870853f141b8fa351","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"f93fc4234d77448c8c8b72f7146c22dbdf4b656d9094d0399c5db97a7ad31e91","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"023108c90068b41af9bb580cfdaa75450fe62fd8a97996715b9f11bd6dd2339c","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment.json","sha256":"b448e2ff155ee9b312ff12bd3dc11b94fa401b361a5d86be9b1bd12036e85b70","bytes":5340}},"requestSha256":"8f0066424c83091c7d2f8744a9cc4d33eb71df9c5aa805f6fcf0234dd57cad6e","metricsSha256":"ae1c9743233aafb9a597ff8b8f993ca8fb4d3b32b7866e64a4073af7f9432014","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":20,"attempt":1,"job_id":"2479668","recipe_identity":"f4e2c782f1bdceb4fdd606b0dc0f5506e7d46d28d7d048898a248bcbb5e6daca","loss":3.6382752088299606,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"5156b6ce9937c18bb2ed1e4ea278686cf7e042bd071560e51a0824e290bd060b","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"25936ee4c73666f3d17ea8525f00e45b9fb7102a9c44361b239a9d3725283d20","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"3042140d4ac9ebf4785dabb0e23ab4d328e21090dbff95a1c6e70b11ee354745","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"07acaf69450903ab810218d0750e7de0c0bfbaaf1a01b4e6e669967f688cf567","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"c8adf2dbad6fa496a6aaeaa3f08c768e719abf99b0304e357ce6b8ed32a78af2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"197c4a3074fe94d1892e32f09cea2538caac69bc83c29b562426e2666d42ed2b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"d6dc9340c607ff52eb807f75b552e05b3c29001d9b51c472e055d0930837ac77","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"1682a01a89809bd230bc5b389493621030344841cbbdf258a91143ddd98489a8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"8b0bbeafcddb21aa43f7f8e367869a72f207dd0215ac493766014cf8ae3848f6","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"65cf9a67a6840df6b5ed1225735d1a13cbe04d2cfa889777d7d6a60b7f2ac635","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"f8d05ab83496bdf0bc1538535951e6937f5372cca72254660be36af26fb2c35f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"5a10c222624f919352caf2c5954e1510ceaba6fd39d42e23d0b1d56f156bb8e7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"1a0243c1b5076ad55d51be62443edf2c3af9700451bb2096f05b2c86f8bdaa36","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"7571e66ea5124b1ae8a7abfb32583567612698ae546a40bf62a1efa22fa20fbf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"4083ed3d6ae100a0eb494e3eeac92db342dd9c9494138d468965ef9270406f1f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"b1cd3d0a1ddba2b4625445dfb43f82a22d0f001b076b8d7a2ad9da727c688cf1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"73507de22070b4767700c643d89e7b53cafd1f4328a715e5e12b8c1a0437297d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"154fef8ce2434de3f7685a2ec4fcd7a7c42ee3f3a731e79044212f0e220a6707","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"d6b385acf7f8088246d6e334a7339bb43e1640c826961b829a08ceb9ae7389a3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"78d575288b63f81a0fda1a84a29b1fd08afdf7e38e4b1e9af5db15e60871ca4f","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment.json","sha256":"c5a8a19b46185a0599cbfff94be029693efb9318b91895c33486e0f243e3b6c9","bytes":5312}},"requestSha256":"c4ef570673f98418ed015539bad4e49c6cf2f12b620158149914abac7b6ba7eb","metricsSha256":"0c1486e301a1b916af4104951dcf5875a9ec0061b173f880544a7ad49bcb9512","continuationSha256":null,"continuation":null},"awc8-wsd-h20--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":20,"attempt":1,"job_id":"2479669","recipe_identity":"1b7211e3f29624e396b9c48fd1875bfabb6286cfd260c8af322367fd030273dd","loss":3.640122381938774,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"a01c15b0deefff61363ca01a56fa3b333b312414f258d4edecb0a9691513e265","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"c799f84196901dbad9e040b1ebb4658923c76a61515e2cc034397171fa59880d","bytes":1958929237},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"67b334dbf577524f5e808e0d836820206e5cd9f599e174fc87760b404c9337b8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"2d2c5c5107bdc8be16183d95ddf29e0c313a2d96a239ef4511e391a800f74ffa","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"c20750c4f05eaf55feb7b905de30906faa97e4fee135bc844c0659a1a15e627c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"96c8731e850f2830ed79f7cd010bbcbb3af4c8ca9c8de624ba57482181642547","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"2f903e6357fc1adb6d4db9fa22a1f8424de5e83330e49cc665818ba24c830035","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"e2a615d838060af87e8c38a2b641014b597c22d30bd22d800215b1f44bf473e0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"d31bd26ddb10c2fd8fa6c3d71d416a2f47d96a02ee2f1baebebcfd2b5cb456fd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"eec8f09cd6da935d8e3f9f83e8da57f0beb461f384eca43674abcc75b6ba6c36","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"4dd8a690e45a9bee11974d679ec8283c693a394939b9ed0c4212fc7991ffebfd","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"21f6d926b27950e1ac2dab4ae1aa70db6396a3aac040dc0d552e5ef60b43b53e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"02cfc3054480e807c5a6d646a1a52fc3c201d0e665fc446aa72f48b04c66165f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"ab0daff2844253c486411063a50fdc243020a4e4ceecf3e1a10892fc321354f9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"7b01d563655c4163fb1f10211844ee1d73c906ed534fc894eaed24664bbe280f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"04e6f01899d65e6e23141b33045bd36a9002c64a7b0225906dded750695d2538","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"975aa2c3e936dd6fe5f036b58b2605484d263a348dfd3756d374927751160038","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"a583209060ee79a24dede23ec82e5b2e5defbe60d1e2ad1fb0a294508d20b755","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"e80116c26109538136153681d00edcd22114dfc8df45e62919cd0d93f434adc2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"a73e85b10df2ff680604ef211baee647e57fb6e4ab6bd921791897247ff6f7da","bytes":244926383}]},"parent":null,"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment.json","sha256":"90a54cd0a8435be34966d46b0a35b18f354fe85ea5ed7517ba0bda4e4ab90a6a","bytes":5312}},"requestSha256":"034b57de958f27a2c062f898cbaa4d0d8e4f6a5d8fc8a2454b4bfb49aeffad3f","metricsSha256":"4ce4d64c61452502879f39c6fbf5941ea29cc056bcb6043884eb62e1ce1c0b5c","continuationSha256":null,"continuation":null},"awc8-wsd-h40--0.006--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s42","horizon":40,"attempt":1,"job_id":"2480625","recipe_identity":"a320bd6bc38c54e847c3074dcfee6a7b5c49d978afb5ce7eb90ac0574d9dd497","loss":3.6209855104109328,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"55cc53fb5e39ca4dd2f6ab2196d62c639882530b9db6e7a050ee81cc448393a2","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"36353ca3da26f77181487bc18590b5688a3eacc4076304917b1791e61e18b70c","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"a5342189f4fbccc8084155e64441f957c23a7b83bde27a4b18c5dbff74f339cc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"b69635b53bcd3fe7827e3b1ce8a47aabc4fd0c13db242a3a460a9828b6907e62","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"c1b623965a4adfd42256e70a6da07c4ca15b6d5a8506e8e116632b5fb19b823e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"f5165e2cea844ae00d1e10c59a7155e23bf64cefdb32d633c308abc97d15c6f7","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"3f82fef1d7845d7755deebb5f2b7389a24dca226f79cfe5c558363919297a116","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"cb3da5059bfed61e0980a7cdac4e714690af15f582d520d8b7130f74d1e738c8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"76a2aaf137ed0e740c6a7e1c6d1d8f172f1a43a1e097b6bb58d888bc1e652b80","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"77e1b217589879a0c443166cc7feafd2846cc816c0c8e98768dc8aaf91c96b54","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"60184a69275eec587cf97c90ebfcc6ce0cad3d947b7bedc562f85c644f7bb4f5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"0043dff9cb4b339f9cd1931508abafd16eae050138376757ab6ff3a1bad92e0a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"ac7903ae2aa6728d3f8a1f43fcec3a882417070bad5401677a529d958f4fd621","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"fb3ba1b3a88b67de804b0c5acfb14d686e8694ba7d88b4c56a290f603938c1b8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"d179cfe306630bd8659f3719913e57bfbe611007a9154abc223c222c4b0971fc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"17c4ab9579fdf615629b8d8cb7f562a22c0ed7d8f191f6c6c5242d3319fecd33","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"b0ace57589c17730d5bf2b46d23bc74c75995ee3d20d61c90bbb9079bec59b6b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"9a74a4fced0377185ed9404a690876907446a19a9411d025f73b9e39692e3a11","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"61e144d90824a1cc2f2369267b3628fdcd819e3415741c7059d52cccc414f05b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"956f7dc0e15107323edc2beb73522ea4312d9868ad2cf467af8223debaf9370a","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"7497b9b5a572395ad291d1bcdd5441bc3cd0487f2ef97de461803dac274e9173","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"f53c16354670adf826a65353c9d624d8fc35e8aa7d019a89d40ce3a2446dddd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"3a4183347a2dd3af7fa3da53cbee917d6ab5372fe654cb8c92f4e94ce03ff1d5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"423345be27b8fbbca500e336f476fc637400607d63742094827e05b49ae943f1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"c5feb1b40580bf3e4b9d510824e7a30b5546270e187a741b37b44dbcfdd2cea8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"4985b79b803edb24d37544eaa8b75e4f0b4df49e3e4350e9a1d0b6e70fbb4513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"19358ed29d3ed5aae267a685e2b9cfcc5f44bac909c6139ae40819491b8632a8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"3e5a5145797ec7c792c43ca493559e00ce2a9e5e8f46edb1068fb5bbfa451125","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"f05bbe00615c46dad0858094d7904a3e3c44b3fbb9bbf4583a7c7909e3264065","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"de71023a2442f3083edcd7b6371b4a5fbeb8abae0a86bf421ea06bfad7eab5e6","bytes":5309}},"requestSha256":"3f009a570d4da1b67bca880a075591a71c59e07f57f4ddba6df13bea35dc283b","metricsSha256":"3094232e914e6e1419ad75591169d74158896eb1c9194dd2262bc78a28449281","continuationSha256":"3c2a44375f6ac2bbd4f19e5140ae95435a69eed9adce1e5d72f1fd03f9dd0f4f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"7497b9b5a572395ad291d1bcdd5441bc3cd0487f2ef97de461803dac274e9173","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"f53c16354670adf826a65353c9d624d8fc35e8aa7d019a89d40ce3a2446dddd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"3a4183347a2dd3af7fa3da53cbee917d6ab5372fe654cb8c92f4e94ce03ff1d5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"423345be27b8fbbca500e336f476fc637400607d63742094827e05b49ae943f1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"c5feb1b40580bf3e4b9d510824e7a30b5546270e187a741b37b44dbcfdd2cea8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"4985b79b803edb24d37544eaa8b75e4f0b4df49e3e4350e9a1d0b6e70fbb4513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"19358ed29d3ed5aae267a685e2b9cfcc5f44bac909c6139ae40819491b8632a8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"3e5a5145797ec7c792c43ca493559e00ce2a9e5e8f46edb1068fb5bbfa451125","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"f05bbe00615c46dad0858094d7904a3e3c44b3fbb9bbf4583a7c7909e3264065","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","sha256":"e0d4731c8623743fed79f494a37b29f38d31496d284ac631f6a3e589b35aa8c3","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fbe6baad5480f18d5b18edd60930c22d58698fe4e567d2df10160d9e4e57708a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a320bd6bc38c54e847c3074dcfee6a7b5c49d978afb5ce7eb90ac0574d9dd497","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s43","horizon":40,"attempt":1,"job_id":"2480626","recipe_identity":"226e88acfb25e532f321fb3f5aa910cf80f1f0f25cbcf82ee3f0f281ded3a2ba","loss":3.6323833602012727,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"c7d1363042475d93fa4a6eb1a7c7cb718ef32fc2e9ae6b317ad120c93afe38ac","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"2add019157556773967fbf4a5eb4ae6d48f8b584dbc5cfe07c3c984af9602273","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"95a3ea2542b22eef59c26fb45f0281deeb5982eb57cc44dae01f0488a0c5d98d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"3db521b3b28681524a873e325a6c37fb5140b3d4c2de889be009475a8d8a54a9","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"c35fed6e182ca5c85e752e32634d56eaf28f17388289f24427e81c33c54c636d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"59c623a32fa18ae5407a5a675a83c5bf506a73052a8a45afe2c1e964cde477e9","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"c9992a21c7e7c9be8e9980921983cf216aca781a654433932e6e8c5b9bed8868","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"aed6a1002de651dde90f1914d7197d04de97b345154ba6bb4bd81ae079b90b23","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"42805d69763e2910c724cf3b1fb6d64b9e6687996eca5a0e1938b596b7fe25df","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"99166faa22f5ddc37b8700fc3990de9fd4b3c321e0d4d38a987666b176080593","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"cb1ca247dc5bdfb91b498bca0df11bbcb8ba7d1c2f4746613302365ecf342ce3","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"3b89ba470428dac7d48cd156e77c14d8fe2cdb2ce372098631a7ac51fdf58fec","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"772724a421fdbe4eabad5fd392b7f9a3e178d5bdeaa1686188bd7a0e90ebc1d8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"ba32d5d341445c6e49e91da99f7d9107fa232dab4b8d39fe0145875e2474f847","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"88487699f2305748d17e2635e14043db32efdf44fac8e2c1b068a93e8d9ef3ca","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"5111ced37031b55c4a099724577dc729ddf2e13cc1d09dd99ae50f79e127ef25","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"7968d740e8382474538e95f35246b36fc63bdce6874be4274c3b072ee853989e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"1683338eeb1d01c54ed10651024f4d3d65e2806ffd6d1d75aeeb2eca0fd10e1b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"ca0daee44a57f68cfb67781e87d40ede9c0c7a8b660fe76e21986acd5e2fb1f5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"540d16a81656db45f34a9efec7003227c6cbaff8104d36bf056d6f636d453b89","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"75be3461f98df19dd062f7603c0d0fddb64d238326a063e04658df567b34fd6d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"ee8de29472447441c4fbfcbef0d27c0d8898226097d5dfe2f9bd6051efd7dcc0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"2d5dd8de84b67667369a5a756e45b2bbc1052528c09a97623e63dd28f2710d17","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"91ae3ad007e7c0b93955a743ad30eee363ac641007d2a656d3e862ee160f61b4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"8e2d06e2eb14ecb65557991ace2c112ac150b01caa774da49f5203b95bf05ddb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"e02860d0e7aa2efdea24015bc8b7046a8070b997e4a9c9dd8b79da285b419723","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"699f5314de1628a83927e56f3ba169bbc47a5f99a4ce9294bf80ec900470ab22","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"1ab768d3e1592f2abebb6795902a00f1fcc3f9ad3692284e2b39c94534df6ddc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"046b8e92221b0277b6d591f15a4c5f4174af67dd15019ca2016112b0f1cfeb70","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"dd64a9d0b1bf3f627d7dee3c26ade585947401cd8552ea6ed0c01dd25e4ca51f","bytes":5255}},"requestSha256":"8bc38524ca4d13da3ad1c377396d341e7f112689bcddc91e459d2c3c7705398b","metricsSha256":"57b137e2b459fec7dc3fd2a4113567e06be246e4aaf22d0dad81afba1aeb1081","continuationSha256":"fac713a050b0744763605151fc5fd1a90a94d7bc6f6a9452500fb10a3782df1a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"75be3461f98df19dd062f7603c0d0fddb64d238326a063e04658df567b34fd6d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"ee8de29472447441c4fbfcbef0d27c0d8898226097d5dfe2f9bd6051efd7dcc0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"2d5dd8de84b67667369a5a756e45b2bbc1052528c09a97623e63dd28f2710d17","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"91ae3ad007e7c0b93955a743ad30eee363ac641007d2a656d3e862ee160f61b4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"8e2d06e2eb14ecb65557991ace2c112ac150b01caa774da49f5203b95bf05ddb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"e02860d0e7aa2efdea24015bc8b7046a8070b997e4a9c9dd8b79da285b419723","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"699f5314de1628a83927e56f3ba169bbc47a5f99a4ce9294bf80ec900470ab22","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"1ab768d3e1592f2abebb6795902a00f1fcc3f9ad3692284e2b39c94534df6ddc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"046b8e92221b0277b6d591f15a4c5f4174af67dd15019ca2016112b0f1cfeb70","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","sha256":"d188f9eac327ed7da4d733457954a93dd2e804e6b6f2a758c4410dc28ccc69b2","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b4c77120ba532ea4f6cf8c3483edab4f9ebffa4d896debcb92476ab831d0415e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"226e88acfb25e532f321fb3f5aa910cf80f1f0f25cbcf82ee3f0f281ded3a2ba","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s44","horizon":40,"attempt":1,"job_id":"2480627","recipe_identity":"14cfe52532738e879450ff01defeb4d491386b29c1589e5d19e5896da721a759","loss":3.6347050611070006,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"201597466170baf22e7abf990c7def43e29b9633a690687d98e312cbbfa93048","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"79fcf7e243fc32a4c6555b459df2701de177e813bf1eb25d7bd1aa2cd9984901","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"1e5d88a74f7e9edb3003a7034af15f43d0ed11a6d34d77173a9e644fe8ccfdb8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"9dd91e27ed7876d5aee1bd81eb109272fe04043854eca0cfafb29b155ca8e275","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"aebf0f794ca6edb7bc57ee711366d6caece7155ec36f39a80d90fff8b259d457","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"2dda80971ea8359417276cec1e4758da296f1e4874e9144776cf4aade24f6f8a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"46b5b78682de1aa92ff02b02b446484da12aa2c035f896e7dcaa023445b326b7","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"4402aac396af10df3811a93d3aee5c220f0a86f9e62e44350a81e432c224afa2","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"75e7df0524620a3180ea650e345e1a15c3c92dee6e7331f1747061fb24695d7c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"8a5a397d6f3eb0b897eb68c20a7a42c82f354c8c6e75782af0753aa51f5e0597","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"354b1068e55e1bf77092bd76ccff15e329b6f9b8b06a74a2f6ee3d59e8228f36","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"e9d639fd44e11153a446aafe78eb33ed443e775069037cd4684294717314ace7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"683527d26c2050047416e2da913f24cef24b94348907cea603e1c7da229801f4","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"7f48324975ad1ffdf6b07c942f81b38bbded38c481cabbd38efc1f1ee42f7766","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"02a5a676a11b0f38a5b80fd3007d6d183452a72dcd65c311e55b91ed744846db","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"80528f822202c63b7b42d66a3753b17691569b022d34c22965320dd91bca6f4e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"4bf4c04a134b1c89f6eead661a666cedf08788cc57ceb21f7490372bcd97195e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"b994b9841755f63409631365bd97433329c15495c24b8ae00ec736f2a4b0b222","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"99831a08373b7710493ff7b63d3a58552cabf47d0a562393270cb4d96a89d024","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"728c3ed4776f7cdfb99d9e10ee908cb01b97a64eb5abad626b8e68ab24f01a1c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"384571975bf8f37f09a1a3f562d4bcdbdd35c16e26337d83f54fa667a3e98f62","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"9179185d8835d177862fe8245edede3e9a7de150b6666e35f9e4f40592c20b31","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"6bd1b08df5dda05bcb43ea7aea2fd9bd4b2025831c6d4ff003840704bf50403b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"ff6d7586aed5d99fd250b41c8bf45c0f6b8a3f8b93b8cd4f280c0f21940d7231","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"7a2993088e182539fbfc2d2af5316bcf6393ec99b5925ccb47edad4529870a07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"df1a42d2f780abcb90b88ffef452bb05d48c50309ea6ce44d3b333b615b67ad9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"41de6361c17315983c1dcabe9683295ae3142c87a1230f8b1ef59098c663ce99","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"d029d1c1d8fca65e872d220e492b9a9265527c81ca21779e69ce9388798f8d48","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"86ba63b49c11f65ec3a618f90d1e0e2fefcfe7c4dff46f9997c9a85202771ca7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"d287e1858deba0aa5e9905637cb013d8d6db9c0174a00aed8170fb85b48e9f20","bytes":5339}},"requestSha256":"67082bc24fad05ebeb3fd01f86b81bee1907063c0df9f56a798bb2c6a2efd2c4","metricsSha256":"51b6878fe0f7f539a90768dabb38ee00d7e2aa38995236ee8b37191c085d8ce5","continuationSha256":"7e1bbb28e73328a9bfebe2ed91e63fc75aacdc53cda3bdbf66f1661dec1ba9ee","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"384571975bf8f37f09a1a3f562d4bcdbdd35c16e26337d83f54fa667a3e98f62","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"9179185d8835d177862fe8245edede3e9a7de150b6666e35f9e4f40592c20b31","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"6bd1b08df5dda05bcb43ea7aea2fd9bd4b2025831c6d4ff003840704bf50403b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"ff6d7586aed5d99fd250b41c8bf45c0f6b8a3f8b93b8cd4f280c0f21940d7231","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"7a2993088e182539fbfc2d2af5316bcf6393ec99b5925ccb47edad4529870a07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"df1a42d2f780abcb90b88ffef452bb05d48c50309ea6ce44d3b333b615b67ad9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"41de6361c17315983c1dcabe9683295ae3142c87a1230f8b1ef59098c663ce99","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"d029d1c1d8fca65e872d220e492b9a9265527c81ca21779e69ce9388798f8d48","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"86ba63b49c11f65ec3a618f90d1e0e2fefcfe7c4dff46f9997c9a85202771ca7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","sha256":"21e315848c8ee611eb3ff67649e3fdeb21904d156258d52dc438f88d8061167e","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"644f24cce52cf9030794a8fed5b953ef7db5f4553d037db2b688fecd23484a62","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"14cfe52532738e879450ff01defeb4d491386b29c1589e5d19e5896da721a759","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2480628","recipe_identity":"6c946f13bdb7d58068672c95356b77c1e0a2a13a8eedd488308a7a04ea6a9a6c","loss":3.648053087155755,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"ab0b691d6c7eb8f4a2906fa438d88127a3b9c795824e41269b521fd4c35d7f9d","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"a871a2a246dc6be004e0850b32be3efdc7d95690bc1cee26603febc08cf66c06","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"afaf03613ebbc0614ec7ccad0ce4dcdda7d8a9538bd4de804f0deb48c6e43e60","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"4767dc2972aff08c9ae3a9e9d19baed0737daf0f838cadca2b6276d330eb1dd1","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"69b7d3d0c7579575d16f3819560f677ae22487ff8dee34f004da78cf94c18dde","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"507a71e38b50a6ce054e73ad05990e7b80fc7ff873d02c48dfa18773a777dc1a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"63d0b92c468a5392dedfdd8eae1aba89fe0b2f1882146bf6c18871c37e718662","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"296543e3bdbc74ae6ea4e458e9f452acb4da26fc95c7e2abf60c1d5f37a721a5","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"68956ba0fde2693c9a8c32594eb3875bb944dd2195e949a71d12661183025842","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"0e6ef0bd80fcbea7d95f638837024a9fbf068a36fb75881c4865c6210ec7cee9","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"f6a50065736ba3c6545a51721904713c57d4e677c90a8dd6ca249e7c48502fc5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"b6bec49b13d8ce9425a129c49711c09e4daf6d4213f9f1be0b03b738aca3ff34","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"68ba61ec9cff8355cc4d7061ca41961fef8055036d9ef20710b84cf8649a8d92","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"05df48b7bd6ec13e7641c167c994bf976b409d7dfcf4c8272de538f70a63171b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"96aca5e3586052d16ac3b8e97c9fbe010c5d833a1baaa7962bbdff4dbb3248ef","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"6d57af0ca127dfb6162457fbcaab3bd6601ef778e4a6914acd6b49800c7f9920","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"ab19aa1d6210e10c0ef67e58387ee31d37c165f15f38c936dcb49ea4794656eb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"030062b83b9f1d34c8a1fff22ea7d28f5200fc24069bd4c84e96592b8ed38b13","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"98fe7c4afeba872a76f247384363d20b050bf39fb0f36d14b0fa3c7c89d39e07","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"511270b9f23d1c207ec2b7d1243af01c9d942f5d6c55332f8aa4c24fdfb9df14","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"9491fd223687c9d9c71c65325084cd8c7b8ceeed38381b767403b3ab3e9f3507","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"1c5fdeaa0cc1dfca96d42d843a485f0fb0b26d18b8e85547fef20eebbe9c06a0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"53c48df9cff2b6462fcf5bc0237bc90a080cfb09f50b31005b055e476783800a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"3f6b73ecd746928e167003cbec905ef4d6f1a7fa074695cdb8840ca946a62a95","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"07c66f559e0f5a46ec51a4f44ab42882c45462dd3e88522a69c74f2f338cc1f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"3e26f32a49a1286b05b892a3dbaedef3e1a7e62c7bbdc20c5f35aa26584f28f7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"ad7a26134b22da21199f9b9525bd1b3e5d7707642df5504dc589fdff17107435","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"baf75e43c0e6ffbefc15b0d68d1c56df81cd1d44f65b40c2f62c3cd7b20cc6cf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"426451b0c3aeb7bf70178f306016ae6de11cdb58c2ee3d49599771d30db8167b","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"2fa361a5c7362d924e458c099cf5a1c096b08ec84e64147138783c459ceebc54","bytes":5253}},"requestSha256":"e2b29963dbeeb2aa96aa9205088d27478abf1949d2f7c0ced4dae8718196631e","metricsSha256":"e3c6cab309ab2f9cda6a8de1ea56851f4706967060ab949ffe33d7226efc5f80","continuationSha256":"fb3c2edd5fc2618cf4910edfdeb7d8b8b6d70750535c95df7e9a3804f146960e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"9491fd223687c9d9c71c65325084cd8c7b8ceeed38381b767403b3ab3e9f3507","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"1c5fdeaa0cc1dfca96d42d843a485f0fb0b26d18b8e85547fef20eebbe9c06a0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"53c48df9cff2b6462fcf5bc0237bc90a080cfb09f50b31005b055e476783800a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"3f6b73ecd746928e167003cbec905ef4d6f1a7fa074695cdb8840ca946a62a95","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"07c66f559e0f5a46ec51a4f44ab42882c45462dd3e88522a69c74f2f338cc1f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"3e26f32a49a1286b05b892a3dbaedef3e1a7e62c7bbdc20c5f35aa26584f28f7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"ad7a26134b22da21199f9b9525bd1b3e5d7707642df5504dc589fdff17107435","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"baf75e43c0e6ffbefc15b0d68d1c56df81cd1d44f65b40c2f62c3cd7b20cc6cf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"426451b0c3aeb7bf70178f306016ae6de11cdb58c2ee3d49599771d30db8167b","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"e59a58608d2a949e2298db7fd0b3c881da657a9791cc0550a7b354862ea7bdfb","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"164f329e060f85ef8de53bc26fc2eff9f1d2f6e968a366edec608ea4ade3a681","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6c946f13bdb7d58068672c95356b77c1e0a2a13a8eedd488308a7a04ea6a9a6c","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2480629","recipe_identity":"863ce256468e42d5d33283d28a961a886f4f12f77c122368c4c24e767fbf28f6","loss":3.64414889483402,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"43e25f2e4e90fe69d0ef43b4de0b1ac0d5fb8864a7e3dc7d5f4ab621809a0638","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"7845d19960c46c193978606ab941fb13094ec2ab09f9bef1f8e46f91f6387935","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"15b7cc2f3e439ce74ebf1b30d2a134703f155ab17dd6c30bf65ce952e183f233","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"2b42f660758784629192ed9db5fca8a666720f9850d6744bef03ca8be1032577","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"cac29ba0bbba175b5211431a443ce6659da6c8add3758c4d53599debfc8ff19d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"78444e497115576520f7ff61697edb3091e067cbb3027cc6e92e44528b122359","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"e0dfe3f90054d44fb7f8316cb999ddb9588bb2f33a4296b99fa97ba025e2e4f4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"210f483c7995ac99d432ad1da5150a1ca7eb9feeeea388e5389c2b2aa360c41f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"ebd5864f2ca5c23f040f64848f1ba48faefc904b7959a1ae9f30a3de1b041ee2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"342b1c8cfa1fb9a6547104147d60636e8c3b120a4bd1952be176b5318c53f181","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"d93e1b521d4de7154e3586c65d726a0384f05f67fd94954c6deb713c1fab1370","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"c128f03465ea7c9644f5ae3b1f748e1945fde778001b2c0406f6b8f468d410d0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"36ebdb6c4a7339a30ee5323d96dd0d5e18d76ea4f567dbf8ad65a5f319078467","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab8364d880fa109782c8a6d4327d03c106df4dd695d04d03aa40bc903f4461fb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"dcb248a6167001e3aa1fb4cbaaa680a64a578ba3ea5eb5e68edc1fc4bac6a9e6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"d4f523f44c5df3bfd017043c216719a368c0b06ab1f9d6982860fa802d2f19a3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"e005bff2f8ccc052bd26992ae2d6644f5a774737f24d6b80d83e1a7f4a260a65","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"269575e26d8da1cf5c63cd2488734521d7f463de07f1fe7aaf156bcce24797d7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"bb4add6681633c178fd3348437d7e386a54c9ff7f695cfa4246d13fcd11ce0fc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"e7a51ed8930518698a489ab609b46bc77d8d551f016b70f0f2efc4805e9b379f","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"874a0e756cc6afbdf46ff512c1a24552b423b0b7f536bd6efa53087f1f3e2f0f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"cc61d0126957c668e2c393389c5f8af40887c02ff434adddb07297b6d8044379","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"9cbc68e0e49c73f2d470606ebc464e39297d6583d6901ab72eb2649be677a826","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"44133824b729aa5b08243d26ebbcf03110077e7a75db0c01eb1e5ef9b2e1d68d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"0208e616d991566e26dbb8b7da16314370aed855046abd412abbb60b1c841563","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"aacf65d52cd8c251553d3049977baf2f8e8c63eab5e43f3b5b0eed018f6e0d29","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"2264359d0ee4695fa85a9b80fb0cbffddb6c28b087250d0ee4077281f60b1c5c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"f215299ff5f1c8a6d7d10447cead981b121cec2ad426b3e14946ff9b948cc437","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"437ad99a9a2ecb96a30c5c6a3300e30513a710a52f9115570c5b007a966ca341","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"be93db672ee0c2dcdcc3c9bf38fd7c59228f9e7753f53c28b39eed75f11e33fd","bytes":5253}},"requestSha256":"8d7a268c69e51fec97e5116f0de9f19c51a485ac36ba6e58507744f137132e2f","metricsSha256":"0ed414385b161fc86bb849f9b7d2bde805fc7e972b8a1f8ee04af012022dde37","continuationSha256":"179fda24832019b950f295fe55ffafe9456fd73b36dc2bb78fb9d853817169b4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"874a0e756cc6afbdf46ff512c1a24552b423b0b7f536bd6efa53087f1f3e2f0f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"cc61d0126957c668e2c393389c5f8af40887c02ff434adddb07297b6d8044379","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"9cbc68e0e49c73f2d470606ebc464e39297d6583d6901ab72eb2649be677a826","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"44133824b729aa5b08243d26ebbcf03110077e7a75db0c01eb1e5ef9b2e1d68d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"0208e616d991566e26dbb8b7da16314370aed855046abd412abbb60b1c841563","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"aacf65d52cd8c251553d3049977baf2f8e8c63eab5e43f3b5b0eed018f6e0d29","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"2264359d0ee4695fa85a9b80fb0cbffddb6c28b087250d0ee4077281f60b1c5c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"f215299ff5f1c8a6d7d10447cead981b121cec2ad426b3e14946ff9b948cc437","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"437ad99a9a2ecb96a30c5c6a3300e30513a710a52f9115570c5b007a966ca341","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"4e6f89898acec9844f8f044b4be988c495d83d93bbeb0ba69ad48108a5998ee0","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e7f152d5c8309b386b1459fd261dd80984b2e568bc3f67fcab3cf959799f123e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"863ce256468e42d5d33283d28a961a886f4f12f77c122368c4c24e767fbf28f6","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2480630","recipe_identity":"05ffdc6f00a8cf2b69a2cccf11ba828b2a92b0a144b707559359e67e3b519660","loss":3.6421183747577417,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"8996364c6d2e86abc25231fdb36d4a0c9a14dd0e6909fece47b3bc371b997e64","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"b4ba35959649a3ebd3aece2271c469ad4a65bf33772d660de3eb15adea821ce3","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"2c866c3fbbe8583c101c8c75c3ea580ea40d144bae5a68877f1c97d5057b49e0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"5c0d2d0ed672644ed6d4e2e62287474880315eb9d6800a09dacfd571ab4d9c94","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"8dee7175450fa12263050a9360ec5500a422abb3486fafe0ef0102c459cc2a14","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"80259a42cbf5a6a675fe715171c873776a3fb82504124e5f5a88cf30d432fada","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"b9e4d6f2535561155a2fe8671adcf9569350191e32ea4a72226273349e0b8337","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"adbdd335b0d71e04bc8b98437c7587711d899e0b033fd07d91534ce318f8ed7e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"3a7caf8ebe497e6ad197b898e00345b2eb700e4f2ce6f0b9136e8aa2019811a0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"d2ac7be81f5847a5173cc0cabbe26c5e2de998b0895c8c513a8ee1779a21be21","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"78ecce96f6eb3ce5d6c0a27ad21612760774eb6f5e9c0a7d97f3ec6ae4d534cc","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"49b7cb5e0d0ea908fc574223c7311382f46c9c2f669ae53a7e78863242bbd555","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"3ae44499815b8fd4b8c5f86397c5fc5dff61f0c7729c051dd463da72a2119790","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"250b26dd56fe2352b11b11ca6554a8bfba39db9802b4b27025162503f75eaf83","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"e0f895ba7c957c5272f30aa547d4d5cbf83e75a9f081bac01318e209cf959d87","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"15ddf3987e3f8f1482b6d0785612acaea275c95bc653e1bffd3c1a709f9ece2a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"9e24fddacadd8b1cadbb97f89947eb53f249c68c2c3b50cb72908dfc35febbcd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"b1a35b29a1f14d507fc4bfbc5d672a4c1c8714eac3ff8be8f1a5a57041eab6e9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"a5d2be4a9159ff1d2f5e77fa70d337296db6f8db5b52fb1c17fd18798ee701bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"10762a8cd2e6e8b8ba25df004b091e4b134862628d810821c38113953285db72","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"a306d79ac7ca119ef5fadd240f0ae894d84e331cd30842f2a30ae6efda4b61e6","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"5759bac4a244e725b4ebd50e1b9f3fc04fc082e8ea1f5e60fac13f86400d204f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"cb1939685f8963d67c096be7ba7690fcdac4b5fe27f93aabaf377a703ae3ee53","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"d02bd20097382c1f4c3d774865e18fcf6674ad4aa4265deaf352cda87527e2bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"a50d48bfb2d727bec97b5eb0e7ee1a1530c0fc23941743c967aee4895c334bc7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc46962b5e74b71de4be42204d37fe320e9c1777f1ff5047dffb97ae37832cf7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"aaf42803ec21c0e7939c7b31453a95ef6fcc3bef862604bceadac45f7c2a39fc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"98ce069693d692b663d300d727873dacba66877fb62f788a947df1b28a3805a1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"083cc68cbad58fd5d59b88e55f9bc3b39820081bf3ca2c9d21e888efa6c769b2","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"fcac59feeb623b3c2dd3db8a39718ca85c59a873a2d394758229c1b873e33a9e","bytes":5312}},"requestSha256":"9dd832ae676ba1a00857b6c472c652c555d42b435c75733a83f8466dbe3f3075","metricsSha256":"b68455f6088411d5b710b8de80be2dd99f931c76f644a30e9937068ac3495626","continuationSha256":"dfaeb729b341936986972b3c6b15a590e5753208ba93df76d558010de89d0700","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"a306d79ac7ca119ef5fadd240f0ae894d84e331cd30842f2a30ae6efda4b61e6","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"5759bac4a244e725b4ebd50e1b9f3fc04fc082e8ea1f5e60fac13f86400d204f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"cb1939685f8963d67c096be7ba7690fcdac4b5fe27f93aabaf377a703ae3ee53","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"d02bd20097382c1f4c3d774865e18fcf6674ad4aa4265deaf352cda87527e2bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"a50d48bfb2d727bec97b5eb0e7ee1a1530c0fc23941743c967aee4895c334bc7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc46962b5e74b71de4be42204d37fe320e9c1777f1ff5047dffb97ae37832cf7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"aaf42803ec21c0e7939c7b31453a95ef6fcc3bef862604bceadac45f7c2a39fc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"98ce069693d692b663d300d727873dacba66877fb62f788a947df1b28a3805a1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"083cc68cbad58fd5d59b88e55f9bc3b39820081bf3ca2c9d21e888efa6c769b2","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"61676755e6012453f84a33593ab2215cab15e7ae734cd19225e3be43f3c0a77a","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"96b6b8c88d123eda8edde5f1d04feafc35d3c77bd9a333235a0ed45e0f36baa7","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"05ffdc6f00a8cf2b69a2cccf11ba828b2a92b0a144b707559359e67e3b519660","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2480632","recipe_identity":"a9b11b7ed0ea6407a5d7e5028402cbea7c01167fd5a2b9f86ed23da930774f3f","loss":3.6621054579254735,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"bd60c510e3257a86fa3fc6a1db5b58fa9b2cff7546f7c3da22c0ce0c0e4c7815","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"a872b765f5b894e3e9c104360a6d071fbb61f031b6e21c7ccaa492226dfb7130","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"81ba7fa173e52482e5363dcaa42643dfa5a27cbd1b10b5fea04a9e89a72c60d7","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"6377f5036f22a9573ef932c685de2056f2e4c01eb2b684127fad2be2387c3de6","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"0ba9d79d5082c28e7bd09370c46b39ae7846e98ede04398317b3f48c725567e0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"dfbf4e2a220e7fa883855f335501dfc36a2419286e44847ca266e4e54a28445c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"34b11229e32bf9d9e1a986b1d91b8d7a301b3198c86dbcfbb34f51510322c742","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"e5bd0967c461b0a3508de97004878642d4b6a1830cd0553dac60fd3465730cc4","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"6d958a7c24bc641ab5a40e6d32a0aa73dc8144de001f1c27a9e4b18fc698e3c7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"9d948bf1d50190b885bf34226abc30eecf46973c32ac7759217dbf0bdcb44470","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"96ec4b2253fccc17f0c2f55672dda58e9403e36cdab59c19f0dccd615ac0c77e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"ae80db95877f6fe838c19fc03d3976d4539c13e8af2ce080c9c38d3f267a95f3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"82e8a7ea2eb2bfb7f86c5eb8c354862ff87482b4d36c9023b6cf60cfb83c5a79","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"8e61226639d8426d0efd551ee705dea96e907d7e5f0dc4b1b6c3961baea501a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"81bac4bdda6f2c20ea645479b489aa5dfb41196097e50c1c8b667b8dae4fc5dd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"60ac469ef285ecdfe05ff36e7757b5e03cca76e380bd0e24c4ff0e58c3810b36","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a8ffd806c163dde69e7e063b7975185e72e46c86ca84924fb876870e858645e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"4f194addf19d93da0ab8f838da3f3bcb94fb597f7c522ceb36a4cdd09ba1358a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"cf777c8863e2c6357a704c2154cf0ad7f3b9e092d1b79019b1d6d62263576685","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"0634a5608618ac8c911ecc1d654368bd29d9df849917c51dd30f8dedc4b41d08","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"49a6dfd1bbe63f630fe9facf827ae96e7f100f735170ee604ec26c2543bf05c4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"db5a58760e1bd97d75c214c480d8aa1285834e17993e394c916ceb22968a37a0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"a0844ed215a2766e41836abd11823353dce9214a8b99e3fb20388a1a67a3792d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"f9bd7c08afdbc52b23eec2bea0f42f513ee988870181ba51bd7de99ed51f00ee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"b85791ee88f8461046e33d34a89178b70f5faf65dd21b0f05f122b2001a9657c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"442856376e99da3bf084e94f9a80e297f9c051e01d5c6faba9fb1c52407e46cb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"1e0a88fb4a1a8eb5d17143ce814e079cdd7da1d4c7c59436b8c61046f7743c8c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"2de4748d73d908417a2f1291c0a27bfb34924d18a6599e2c61fb96ccacd50b03","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"4481bb8426398a9f4550f440caea26d9eb229c5a677caccf063df93d13fdc44f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"97ec03b77e483e85d2a7762629734ea1f809a19bca22cc1bc0bff388c79c59a2","bytes":5308}},"requestSha256":"beab40940b46a82b51ea5cd31efc24a8daf1fd0fccfee0aa179bd9c4fd55a128","metricsSha256":"a1af205a89c3afbd9036741e8ced8e7f33f225f19c84cfc0093c99858833ac88","continuationSha256":"973cc216cd408c12bd0088b0326a2b034cf18db8c7c14edbb5039c124cea3cae","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"49a6dfd1bbe63f630fe9facf827ae96e7f100f735170ee604ec26c2543bf05c4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"db5a58760e1bd97d75c214c480d8aa1285834e17993e394c916ceb22968a37a0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"a0844ed215a2766e41836abd11823353dce9214a8b99e3fb20388a1a67a3792d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"f9bd7c08afdbc52b23eec2bea0f42f513ee988870181ba51bd7de99ed51f00ee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"b85791ee88f8461046e33d34a89178b70f5faf65dd21b0f05f122b2001a9657c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"442856376e99da3bf084e94f9a80e297f9c051e01d5c6faba9fb1c52407e46cb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"1e0a88fb4a1a8eb5d17143ce814e079cdd7da1d4c7c59436b8c61046f7743c8c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"2de4748d73d908417a2f1291c0a27bfb34924d18a6599e2c61fb96ccacd50b03","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"4481bb8426398a9f4550f440caea26d9eb229c5a677caccf063df93d13fdc44f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"5f41031f3771898671534c7d0967fd671d41b96ab8b07fee05ca1c4bedb11045","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"505d4ad6fea35b0fdb5d27c120a60d0c39f0ef02613f24408565de13430633ba","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a9b11b7ed0ea6407a5d7e5028402cbea7c01167fd5a2b9f86ed23da930774f3f","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2480633","recipe_identity":"f70909a0deab7b2a89e2803a689f9f1bbc7f23bd7f1a54f587f05584ca0967d0","loss":3.640312957176013,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"169e986a5f99a7935c477c50b7f54da0276714172a6f9027b3726669b94cbbdf","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"2f24f94ca18738e4c132461a219a332bee4f00b254101728ba538f78531b1664","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"3ae2e93b73c83c59f741aef39e9de9f87bb45e432d0dcdab212a15bf4059a76d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"725acec5fbb6f8a708b76a98df874738f82ffb998ad19cceb3198756bf3d82ff","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"ed8a247e7cf1b26741ef0ff187464ab042422f5213a118b7f572b65d73bc8a8d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"acb09e23ea5b2bf430913b40cf88ee12d48a84ecbf9dbf8063b8208b494e2749","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"622e7f05b7f0e4f2a59ed6ca5cd0b78620e3bfaec6da9397c70b75ddaf33fb45","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"106dbbb11dbea3055ca0bb231cc9409a4f98f0d981ce16cb32d2311aabb50e32","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"3d1d18d34f9e6c7235c7bbec2f430eb8fd43d4e0961ca9ba3b30381e31d558d7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"008c10f467525b6f8605c6524605638aa3f7d925cfb42edd85f3c662db2e470e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"56a3639843bc7a03f6ca36bc0005a538e7eeb840e1b73fe01efb923402a16048","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e86310c1d17c22b923975e3878b9cdbd4a28e30fa3909391d163405583eb0893","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"0f6f8bd8d960532ec7f9dab909efd5ea023ff234db498f5f2a6e4c3af3f44c28","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"eed2996a5bf3c0fbba9a651d4861bc3186e5890d4d9ca3758882d5b7dd52cec8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"c8fa62e8110655baff68d8ff1cd298f9a3dd6a0a0d9cf05c60cc35863b1adad3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"df665051f11c58512efff6829b1a2a467dadee4f591f081130f87e0ae834ee08","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"f8d9dbc3751b45416d0dcb99780577c97b4ab0ac71d7b53e7d433b4b84f02135","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"a5e5db1afeb1a08129208221910bf5f2899d15682d1c1437c2543197ac7e0099","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"93e915bcb42685968867d4461da3db34dbba9b26f9fe252fa74437f46ef39b79","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"96e74a35139fcf042e5b0a8a4d72ec42210c9fce087b3c6bce6491fd67da9adc","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"953f4411376d3aaf8f3a7b771f542d54f0ed7e35e4a8bf4372b212337ee5a391","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"3ad82e13e39d13be2a6b490f047e9bd8d4c9e333abc7b7b71d83033ea80c6592","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"2f1d398bd90983e8bbe004236a1f80e29b801132c5222f18fffa318d6d2020ab","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"a1258a961b01fdc084517b034e4d9410f5d612bc9a20bee0e389048faacd8096","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"34080b8a7fb971b5e6388c80bfe89b8762d01bfebd8af1dffe40f91fcede3188","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"0ed0b63bd7ac289f56c5d9673cdb5bea98f1845e185ef43967fb475b5027cad9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"a5a2ab931637383b3bcc1d6bebc34c495066d0c719dce24189cd3c97afdd1526","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"37066bedac40480c32f0e1845cf454488c177c69fac160c9b439a48966ac6cca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"ebb6a591fcfef068b8fc9cf6347fa9205a835c80374e9db057fc4c793b07d98b","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"085331d86e8edf5bed159c7b46adb9e3be8cb97080344662e2d6cacfb0ad5585","bytes":5337}},"requestSha256":"c84ad1fb493cc84375fd84b5f39992816fad494e8c5410497c02d79c5f5ff18b","metricsSha256":"5da3a17ec23b37efececed027ea9617f129b9b9c83c48298b7a64aed263e92eb","continuationSha256":"559d0be4749e4933f57d7f8b28252aea95b2d89ad0d1e2b198e8f3c2f61f612c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"953f4411376d3aaf8f3a7b771f542d54f0ed7e35e4a8bf4372b212337ee5a391","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"3ad82e13e39d13be2a6b490f047e9bd8d4c9e333abc7b7b71d83033ea80c6592","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"2f1d398bd90983e8bbe004236a1f80e29b801132c5222f18fffa318d6d2020ab","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"a1258a961b01fdc084517b034e4d9410f5d612bc9a20bee0e389048faacd8096","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"34080b8a7fb971b5e6388c80bfe89b8762d01bfebd8af1dffe40f91fcede3188","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"0ed0b63bd7ac289f56c5d9673cdb5bea98f1845e185ef43967fb475b5027cad9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"a5a2ab931637383b3bcc1d6bebc34c495066d0c719dce24189cd3c97afdd1526","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"37066bedac40480c32f0e1845cf454488c177c69fac160c9b439a48966ac6cca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"ebb6a591fcfef068b8fc9cf6347fa9205a835c80374e9db057fc4c793b07d98b","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"d4333c48a6a264d59b102e70f4f27d5ad34fd0aaae4aed849358d62a17b3d368","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"126e62c588aced9e33b09ed779a231d459e4f6da554815c361db44a8ee3491a8","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f70909a0deab7b2a89e2803a689f9f1bbc7f23bd7f1a54f587f05584ca0967d0","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2480634","recipe_identity":"738ac68007a8ee6c02ad29b366f5d25fe1096a4f56d7c54cd57737ef6a39b401","loss":3.6506803655939053,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"d86dc852850dc6fe3e2864a1d938ae72e800c09959606244f92c329f68e92067","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"75959b485dbc9f945f80bc422a1c029fe362b9e557580bf45c6e22a61bf13deb","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"c3bb19ac7f7e2b8f95698d71d79242ed7c27c3d4375b72b859d0161f2b460193","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"c22e6322972d4c62072bbebfe991da34e0010a36f3d183a234c8c2f1f423926c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"3e7b82e26d7161dfaaa041f6f05dba157da54cc8457f3ee1df0ab31e21b7e1bd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"ae8bdd4d47aa8087a7aba5b819e19e83488d8b3a72a263bcdc47f2d920a7ce74","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"e6dc0b4c6d73342c042ba992492e87cf069e71fcbf621080df675683fa69c6f3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"bc6dee55b8314fdab12f4abaff5fdcc8c18a316df286186a6e46048dddeea8c3","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"590d84fb86ac6549c9583a9bcffb25200fb9ada4a8067d9801e0d57c291db035","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"f5d40d9261700c86c7dc2ce5e1346d4a9585835abf4e42ede61c5d12223c7e23","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"adb2237b729d62a2b985c702f5b0c55968e3cfe4dab7324596e63103dfd4e8ec","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"3da15c04f0c4e8ee6457edd54d27d778be0ff11cd13ee60abbb301c16d957679","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"7ab01b699e7912f3b886c22f20e9de01f8fc9e2ed2f72aab998f65d926b57056","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"5c9912b05a38b4629e9e8dc8128d60c59fbab85b4f582e5aec8d95f169f44e65","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"a6d1719ae88b6d1abe9dadf362cadc3695b2351c26ea0056aa9b4654d04728b3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"f00930f538d1fd449b85c3c1bc20af2d0787e37c4749fa175dd00866a9f8fda1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"df32e8054f7652547ddddd0ea75c6db79d1a857c4311860e2217999cd1d5a2e7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"729edc911bec74ecd4c7129d312eb18efe2a0c69b292fe1828c85ebb2f537539","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"5c0d6bbe7e96dc5827bcd9b6dd10d14ac257d41eb6a82b8b44220d7b4ed6af63","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"01e44f5405bacda4afa3c04deb96bb5fe393e86ba067053ae9f435d3dac58189","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"2851f62a7cd4ff003b233c4cd9fc20a0bdc96ba036fda072ff1583e2fc1b9423","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"84fca79fc1b0ead4d38f68abbdd4f900337e349e8cf3b374eeee46693348e289","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"0d9deb29232a6eeb9f5cdba4392e0eaf1d243c6d1da29f2103f1b42c286a69be","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"0f652019bebd2fe4c3b728e04e127f0fc97a62b4d1cc8f0507f3fc50b1771549","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"813f5692f716b3c1ac9485fe9ca51abca8f6aee33a1ecc8df9a960a08682e67c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"e5bc0ebda6c766401cb8c9354471b68a04a0a3ee308381b0a0166e4d26ca4cd4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"4d8800c03c35217614ed5055d1bbde38a6d57d4c1cd2f4741962c5d7321c43ed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"434bd8620f9cdc47a9202902b404dce023602a4be37e8d23bc9898545803d245","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"2645bdfb03f6bf226566b8f6ddbd99247f394ad2aa20b545a130c4dcc08c53a1","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"ca26118cb35708bc59a84bec0685839b47ff1bee984028fc0bf03deaf75a1c3c","bytes":5253}},"requestSha256":"4830664ca734dfb9aa7d64d49cc015c76f799e895557372fe2a41b7e963fbd5e","metricsSha256":"2d93ff952c8ef29d9522325fa77bdf00c15a3b02389a9e5f8ceac294ba819e52","continuationSha256":"4e521af1987854a0bb3b6116b61e5ed7c9f4456c28c7f7bffabe73142219ed88","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"2851f62a7cd4ff003b233c4cd9fc20a0bdc96ba036fda072ff1583e2fc1b9423","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"84fca79fc1b0ead4d38f68abbdd4f900337e349e8cf3b374eeee46693348e289","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"0d9deb29232a6eeb9f5cdba4392e0eaf1d243c6d1da29f2103f1b42c286a69be","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"0f652019bebd2fe4c3b728e04e127f0fc97a62b4d1cc8f0507f3fc50b1771549","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"813f5692f716b3c1ac9485fe9ca51abca8f6aee33a1ecc8df9a960a08682e67c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"e5bc0ebda6c766401cb8c9354471b68a04a0a3ee308381b0a0166e4d26ca4cd4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"4d8800c03c35217614ed5055d1bbde38a6d57d4c1cd2f4741962c5d7321c43ed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"434bd8620f9cdc47a9202902b404dce023602a4be37e8d23bc9898545803d245","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"2645bdfb03f6bf226566b8f6ddbd99247f394ad2aa20b545a130c4dcc08c53a1","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"30e969228f3bfec8a94083d9d614d9098aa0ce6bb511e02fcd7877d04ea1750e","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"740e18fdf93eb26f3f127270de2b77e776ddb481af2ad2a935a9c05c720e8c05","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"738ac68007a8ee6c02ad29b366f5d25fe1096a4f56d7c54cd57737ef6a39b401","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2480635","recipe_identity":"7a37cc718c440a30a395a6916856fcb90b91372c0b77c5e62151bbb41405be7e","loss":3.6792063167065248,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"26384bf358ed147e2f9a634dfe41b40683c1206457c068f99cc654671a1eaf69","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"880ba1769557ee4366fb02015e63396d573d2e1f2d8380b95691bcdfd2e56f0c","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"72896414b2e888815e69f0bc1712f0cda5a5abf57cbc8b86d38f93adfea131fb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"47fe0bceffa4db3376bcb9f643b86f7248cab3626d0b8d1155dd6b910ace46c1","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"059b6315eade83c33173092fd72889618ade2e70ef0d323dd761b1e8ddf1410b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"b03836558364bb36ea9fb445887a351dc9f477631729446b2e78b45f99bfc52c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"421d20210c20567d5db7cbc2304a8309e79ab39eede81be42a0bf95afe2a5d31","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"699404b3b6d8ff2c0a8df2d6a167f4abc4bee213051d6c023ffd81f4935618d1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"b82448f6cae20fca0dc94159413c7fd773794fa11e0cc6d5252e38a11cb00981","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"42aa7c50dbb27648b7119ca61f055367c3efc83febbc07c325df30f33a00538e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"d6cd6e7868aae4d440d9d7d6c481ef0f945825cc5be62b46385bf2c0bb2b6b11","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"1b51abc0a24476db23815aa60ff7aa14bb34969b25600c0c054a6c9765470f42","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"1bd23da92329ffc83ab5fff3426bd73318e6e3b70ee393c25bcd336d9786ef88","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"4c36c0b51abeda39bd859e8f2aeb55f7b37cd90381790398f8c9828772f834fa","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"5002ec7996efd313f648ca482d5720970a366154b282a8fb5e668caf335e4bfe","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"a6f794d2f2318240804e4a760ad3a96ee70b95ae7e67c7236580436a1e260613","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"d4caaf2d79d912d3b44ca6e85f6349ce54f2bbcb44b628a7ced0e92f24d1169f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"4d6154acb65784a1fc6ce6aef4b28455018d02f06332cd78843f3a590514d0c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"53f0b6dd4be63646f97c38b8d8555bbd346c673b787c9b5c83c37ca4092c2d65","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"5352160926d0b49ffdb2092f933398acfde7e9cd0e1f0fb156350039bd877a54","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"c3afbd40084ab48a65d9223453ed2d702513509aab4787358849ffd26f4dc11c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"7476859d8b0717888ec2bad9c2523fd314060b07a08c512ae28e3ac492061bc9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"5d364bec1e497ae32148b72ee82bf2cb4cb35fb852462efe6c81b2a437d4893b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"ecd9648684b9222c4a5ff92fe3ae574ac630bb939af7a7fe7b493085dc6a1357","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"7dc39ff7f5122ce58c9d83476ae72f3ccaba54f47441904623db881c8e312b2b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"67f0a88dab993b161ef661c03a536adb1acbf99233bb69334318ca428cbb6cc5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"a9938d6373c5ae7e63b01f21f5261d3221dd9660a6ab0d794a7d657d9d1be278","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"10f9e036a0427986cd1bcfe769ea4c95e94215a9ea05e070329d003555f91c07","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"b7f42d5e700dcb6a71b0968be16d4b85d4f13f29e8b151cc4d20dda735807247","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"57750c80f27841181d1b767e57854ed4d56e3a50f0e2b3d0396b5912e39c2c42","bytes":5310}},"requestSha256":"86235198d15d1fac64487df7d2709cbe8d39541b97fc283a7a6d173ea1a7fb4e","metricsSha256":"1500f77b969ce689e3af5dc3ec205a3cc94099ffa20a228cfeb2edccf6cb0196","continuationSha256":"59935aabfd29067960465f79823c0f140fbf81917a9d03307c6dfe76882fb274","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"c3afbd40084ab48a65d9223453ed2d702513509aab4787358849ffd26f4dc11c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"7476859d8b0717888ec2bad9c2523fd314060b07a08c512ae28e3ac492061bc9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"5d364bec1e497ae32148b72ee82bf2cb4cb35fb852462efe6c81b2a437d4893b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"ecd9648684b9222c4a5ff92fe3ae574ac630bb939af7a7fe7b493085dc6a1357","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"7dc39ff7f5122ce58c9d83476ae72f3ccaba54f47441904623db881c8e312b2b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"67f0a88dab993b161ef661c03a536adb1acbf99233bb69334318ca428cbb6cc5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"a9938d6373c5ae7e63b01f21f5261d3221dd9660a6ab0d794a7d657d9d1be278","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"10f9e036a0427986cd1bcfe769ea4c95e94215a9ea05e070329d003555f91c07","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"b7f42d5e700dcb6a71b0968be16d4b85d4f13f29e8b151cc4d20dda735807247","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"98f2d1dbe33e4ac24eac6e5fe2a60e379efaacbcdc691b63978ac57a31cff3e5","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fbdb252a29c004aea59977f2f7c71c3fac7de8155afa51e4f3a30221f1354e3e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"7a37cc718c440a30a395a6916856fcb90b91372c0b77c5e62151bbb41405be7e","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2480636","recipe_identity":"634ded3f2f4e15474f6f79f49ceb386ee373806292a187a000a35af3500bc1d2","loss":3.6598426073959938,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"71283bc62c84ab1700ab3ad58b781a1eee7f457e643d895a5ecb9b12914f3e46","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"06c69d7b0891bca6de4fa27b80dc342494a24d80694432f55380edaca3513383","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"2cea444af21b644e310378ada76d2dc93cedf4b34d8f1d7be5aa9db4818696b5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"2b97ca779585108fcc3696898da200b1da4267813199fd8836bda0ebc53966d6","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"65c2ba91d33717d06451c94b68768eeeeb8eb50ee3e7f4035858230954d3c5d6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"b1b332c29dae6467821354a79e5dd34d6ee56eaf9b2723ea0af8388c600d3018","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"0d7966c88fec5f513f51bed4d05942018f85debe7dc67555d542b73f4c028654","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"84de29e8b76c1c220233e28534bf166bdc6b020fefdb413f85d317dd9ff7eaeb","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"557e8decf229d006c4f3b837f5b4960aeb11d8fdf657b0053f9dca0d96201ca7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"65526895c6c68a325344b9a23a88339ccb31e21c48a53d038eebbb1e66430bd4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"15b27afaec213d587f19eae2dba01f15a509d96cbd8c33338fdb1fe1e066af8c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"5328404474c6deff30059f178fabc3f298ae18266e35061106c50c6662cc466b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"fc6a931244119a2fa0df9e156b9c7bb84ea56252670e4d6d07feecffed8c76fa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"e0d946fdc261801b738e80303b8992794a0345bf0c3ad906d36591cafd1a0839","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"530bb16844ad6526ce4497836eaff5c954d3a411059656dfa39ce7f42b80c482","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"0bfd36703c1c16db619156045ae7093967da0b57ca6d78aa340d59976afa617b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"1c70ea6dcbc22a55a7a4a4a544c14069f42f4e840c89a59dd2bd6c19be65dad7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"6a37814531d1d6fdc85146737119eb99a14fcb7a640c613fd88da772d220c981","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"51f91022e0c73f8249323947ba85f1dd3f1f81f6f342bcb402ddeba0dad22443","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"0c61ba1bc00c32c7b125c191b7da7a0bb62a31958a907ced20629085c0995ddc","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"81811d4d22627a10cf1661706d5bc180fb2810835b16185e9fab075716e16cee","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"28b97e499d140c51c960251a3a4f7fe898f91ff4f75b84be97baa7b4fac8cc70","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"f3abc7aafbfe5f8a3e32e069827e762fc37d2b598756c0b0bb5bc7dd5bd711bb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"95781084f820dc698d1fb632422873f3280a2b77b39b37dc42499b1d63f107fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"abce995de5db1b6d39996b7f0ac35f0604adb3c8c434394de17e749204f853c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"e36a2fc4934e705e687b4538ff1b0b2b726b681930257feb28cacc953da668f9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"ff272e2acb91037fcff7a54651973514346dbc150dfd9bcc6ea9ad1efbe9b8d5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"f24091489393f4f6d8652c1ce6c3b63119999de55621731d0ed6e70e057a0d9b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"0e1ece2346bae91afffe59f2861f9b99e8386425c2898f149dd45bf10de22a2d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"33a428061424ead177b08b0168cf9e23bdc947dcbd291d93d6b44479c9f6cb21","bytes":5341}},"requestSha256":"0c48f1e7e8e87d7dc7b899caf75dda81874f656d475b68993479f064878e51f9","metricsSha256":"f5d9a8c523f48ebe1484dbf1cf717eea35f824b20fa2c4dc80d71e0aac4c2638","continuationSha256":"3ca1bb65ac67dd1b60ac10361daa083a1a3a9903fa201349450adfdaebd3b6b4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"81811d4d22627a10cf1661706d5bc180fb2810835b16185e9fab075716e16cee","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"28b97e499d140c51c960251a3a4f7fe898f91ff4f75b84be97baa7b4fac8cc70","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"f3abc7aafbfe5f8a3e32e069827e762fc37d2b598756c0b0bb5bc7dd5bd711bb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"95781084f820dc698d1fb632422873f3280a2b77b39b37dc42499b1d63f107fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"abce995de5db1b6d39996b7f0ac35f0604adb3c8c434394de17e749204f853c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"e36a2fc4934e705e687b4538ff1b0b2b726b681930257feb28cacc953da668f9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"ff272e2acb91037fcff7a54651973514346dbc150dfd9bcc6ea9ad1efbe9b8d5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"f24091489393f4f6d8652c1ce6c3b63119999de55621731d0ed6e70e057a0d9b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"0e1ece2346bae91afffe59f2861f9b99e8386425c2898f149dd45bf10de22a2d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"33760870f0f30484730b3c8894cc425f1e07d142a8afe9050bbb8b8f8f77479e","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1c0ddfed5a4208f3b522085649b3af43774c2d5da9919d67e30a699bc064edff","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"634ded3f2f4e15474f6f79f49ceb386ee373806292a187a000a35af3500bc1d2","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2480637","recipe_identity":"e328d55db899d770cc42ef2d7ad0d89d73b6641d9411aaa03965bc32e6db42e7","loss":3.6759053224992413,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"7b36882419b799ebc338e31115e9b12c367248c0fbba84329f11d5736082c2b1","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"5084c8e47f812b81663a395e5decffe3f05a30a9254ca3f1b16d6d4ad65d5fa9","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"e813bbad4cdbf653d4fafcd13f5813df98251f828b847163619dba3518c26945","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"683c1050bec810e267987d53d08d9dadc017193370ed6fb2486323f42393fb95","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"c84524e6cab81fe5b35036880ed69db016774fa21fe1e50907e4effc308c2e14","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"490fc991c30345d1194b6678055e39e8d2c5479478a1a9b55a48cd0506c9da02","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"29010dac337d02639ccf6606b76ddd25f8a56d8f848e43c98d97133e70ad16d8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"decbaf5e1e4d2c70d0bab5c18aad1b184026b06e61925433ac4f8b292810b37e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"65a8ffc6b0c399e826ed11a4ddff57780ce197f98599254bfcf5691957677aba","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"a22b51c7b67feb54aeea2b7ed852c2f7fcf70d8935ebd7372cccc2ebe787f0a2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"46dbc5dabfc15de5c61f990c63e1229ad8fe53b1f132bffbfa2993635bf0f5b6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"de0704404627b554246550b4044bf1948e8edb80909aacaa72eeecf8bccb0766","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"2e17d60c69e0c0fa8bc92592c3db9abfaa7f1c811f14520067bd5456c048dded","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"3d7777093c81aed47da2abd565eee9b0a69297e2051d88ef7978c516af1c4732","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"18679e2a5c515ce9aebd05ab0a7b1511b562ebb184bdb5b3d8024bf4f78089c0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"a9ad27de7569faad0160ba9f3fa8c9dcc2de713e5ec7b8564f2f55221fb82a0d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"5ac1d14488ac283e60aa168ed0f81569b7d757fd7c59e8536e783e4622e43078","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"5a86157a2014e0d58502e9d5dd5f7332d9364e189d25e6b7df4bc69664f114ae","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"73581e97bc177ad49b746db875f069bce88959e5f148e21d74227126067c4c51","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"53141b12dca57400562844c7f7584d06e218e67dc6c411b9994bb32e305ce98a","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"8c07a4c77c29afabbd57f914cb0bca11079071fd297d281b11d9081da5b5e481","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"9d3fb32fab63744000a144827a21e3542912beeb520f1cab54b7610c206d5fd2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"f47174f46fb6858cdf4e8c23f5175aa01a2b1e075c907b15e254b2e03f26b238","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"2bc7540ca9064135fca86a9d750ed86f4fefd80d8d336fb44f6f8ce8ac78622b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"912b4ec512b3d479a1ab89c22dab6b9aa4e7e1fb1ffadf0ec7d4437de1a45f14","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"c612b8bc7c2ecab85a0753c339d98de3a817b7a90efaee28ffd5b96c48fe0f22","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"71bc468db407711eb2d77c093f0f7bd4bc6072a0629f0114e91c0f11c9b3e630","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"37fcda7c788bf1b8b24b4457e4585cd84ba7b8f8880267f08b7d998145cf91e3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"4389b89fa3305d07a21fb88f4500cc67bb3ed1cb00cbfe07ab2c835e0fe40282","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"f53f24c870b6257c8a9bb89f7de5dbae8c904f6f3b29f1ce55a36bc3f6bd43dc","bytes":5255}},"requestSha256":"6a50d3ed4b861a0d06fd1597e01118af00c676bfc2ba85de98f5cc7ad3eb5fc2","metricsSha256":"166453319ddefdbf1f43c85eda8e8649893877b1d0bf5df28dbc7b25e36ef0a9","continuationSha256":"3e18c7ebe41aa05bf5f7338771a5779bb6243fb9c703d0bf2eb6f52f78f389fb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"8c07a4c77c29afabbd57f914cb0bca11079071fd297d281b11d9081da5b5e481","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"9d3fb32fab63744000a144827a21e3542912beeb520f1cab54b7610c206d5fd2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"f47174f46fb6858cdf4e8c23f5175aa01a2b1e075c907b15e254b2e03f26b238","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"2bc7540ca9064135fca86a9d750ed86f4fefd80d8d336fb44f6f8ce8ac78622b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"912b4ec512b3d479a1ab89c22dab6b9aa4e7e1fb1ffadf0ec7d4437de1a45f14","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"c612b8bc7c2ecab85a0753c339d98de3a817b7a90efaee28ffd5b96c48fe0f22","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"71bc468db407711eb2d77c093f0f7bd4bc6072a0629f0114e91c0f11c9b3e630","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"37fcda7c788bf1b8b24b4457e4585cd84ba7b8f8880267f08b7d998145cf91e3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"4389b89fa3305d07a21fb88f4500cc67bb3ed1cb00cbfe07ab2c835e0fe40282","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"fb4dabcf0d9a3d4aaff2a65bf0222401f0ea9b42b99808c646e9049df83609c8","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"057cc957cdee874c0065a2729ebd9cdecf4c4470f182181820a6907dfa54819f","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e328d55db899d770cc42ef2d7ad0d89d73b6641d9411aaa03965bc32e6db42e7","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s42","horizon":40,"attempt":1,"job_id":"2480638","recipe_identity":"b5a45cbfc7efed828568898c64d16ca598b6725dcd40838c76102abaa78365f6","loss":3.5283966212746614,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"2f20ac560c91381abacfb4cdd5a4d5db1f3703ebaf911dedcd9497bb384ca185","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"863ebcda0384603296f867370a62f637776f4edc98eda9e0b9d2ad39952bd2bc","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"a28fba9dcb3a91e6cc83663c135a3b65652dd616d75c486d52b2a37b9cc63a60","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"80310c66426bc0867b725695952845a09e78ee64f7566b189802304be4c1dc36","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"817c261830b63ee5e10fa8f4513a089bd75385dfd709718fb7bc20216e85d85d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"56e41c2f7d7d4a6e127e3746c25d60bf041b0872f7b3c4f11d8888b219648263","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"3001b1ac7732417761dd3706ed3df94c79ccdb2a7846b968520c60b099ac2153","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"300945c049d57d036d58d0cb67f095eea270b628138d3f3c50ff0fe069a38efb","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"b822952d5dc08b422f9e2c8487475a4e8dd95db7b04c424a7dce5145e363238d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"b0ba8226ebbc15af2ec53b576df67267587ba05fad1ad39d0d7a21e2f2bcb22a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"8e41a126bb0c889525d4514d05447e1d8325cef5c36b3bb9f233affa6bbdcc66","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"ac927c8c55a90e48ad8ba2c8986ce3a37c84b797de6d344f3635ec9ccbf9dd45","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"d70438a82afa959c542412fa66e9fa93d68820c0db6c14d9c18db31cde71b11c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"e30491f62636477e5373ce394c63f08747e9e47203d03189958a80e6f3daa888","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"e71f1092fe9bd0d9a6fd891d6134fd4bca768637691d93431e4c0110e3a664b7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"438e1228856b7f4012dbb1dd32c64a14a7b9b3f84527439c48326c8390c3df90","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"29ca9f773d8c3e7eacf7dc256317b9984823ed1febab2ae6851acaad2819dd00","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"5eb3b0a150549ae2d1bfeb0b79afb2a20420f3ac3ca98fd326ebfe120b43fb6b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"59265fec6e4783fda6577923765b91c4ced9cd0abe3121041eb486b9cad22f3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"f1dee93f24c3f82fac54c093a4c411a70f8a2e1400cbbe4436c55957f0b3afec","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"cfabcb9e5f0fcbe105c75c43dca9b56d984089ef3b851f4c2676e94636f5fedb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"9b37e29d973f9b0c7ee6fed3c78615882302ed4d30a73bcdbfbc48aaf67f45f3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"e706439e7feabff4a96ed4b0930c9a191c062c49bcfbf28094421c585aae50a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"9ffa0c4dd14123fd1a496ce40a5f2de4bf280d5ac61b81f321e279d76b4af240","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"9ce2e2be8afc4e257c9f748f061a145a28d58337bb8e46d0c1dabada7c8d2b72","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"102f9d649c402324d90c37f611bfdf421f959ab54fc29a991bae68769fe2c01c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"3dade48c26338f4bdd4680d566e73c17bcd6fd28a9077f75e3bd4bc882edaf17","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"21bfc3a0cbcd0a1d6c5618dc8a7418f2d7cca7e5a0252e4c335fe65761ea76ae","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"da3ca290d076e5ddf464b55a2e203f884f434ce594aea59dc8d52ec03ced5798","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"ec90185f61f0b4664d6428f8b32bb52ff019b3108baf8cbd335b4ee5ced67a66","bytes":5255}},"requestSha256":"c01bba25a38cc66e5652bb9168b5213ba403fc90a9644aff7a58b41c17f7e27e","metricsSha256":"76ab33070b90edb4d22d3ced7ad6240a1823a8b137bbc3de34a26db9a67adefc","continuationSha256":"6e88e01bfb1e58738248787dbf85a9e488e51a3d2fde382256d8eeff8c0eb1ce","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"cfabcb9e5f0fcbe105c75c43dca9b56d984089ef3b851f4c2676e94636f5fedb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"9b37e29d973f9b0c7ee6fed3c78615882302ed4d30a73bcdbfbc48aaf67f45f3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"e706439e7feabff4a96ed4b0930c9a191c062c49bcfbf28094421c585aae50a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"9ffa0c4dd14123fd1a496ce40a5f2de4bf280d5ac61b81f321e279d76b4af240","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"9ce2e2be8afc4e257c9f748f061a145a28d58337bb8e46d0c1dabada7c8d2b72","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"102f9d649c402324d90c37f611bfdf421f959ab54fc29a991bae68769fe2c01c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"3dade48c26338f4bdd4680d566e73c17bcd6fd28a9077f75e3bd4bc882edaf17","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"21bfc3a0cbcd0a1d6c5618dc8a7418f2d7cca7e5a0252e4c335fe65761ea76ae","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"da3ca290d076e5ddf464b55a2e203f884f434ce594aea59dc8d52ec03ced5798","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","sha256":"ab20f677b67152701d1dbf85585e44eb67e36c47addde5dd79847563d6863aaa","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a29152e224b781b11bb46e27222621f1bd112b093d0392208bce0f98de87854a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b5a45cbfc7efed828568898c64d16ca598b6725dcd40838c76102abaa78365f6","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s43","horizon":40,"attempt":1,"job_id":"2480639","recipe_identity":"a5ac5094a6c0730d756cc681f27e85e69943bb37ad7bec445a8e2cdcd6c60859","loss":3.5349761858346045,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"be8fecc079c23b0a597b0102af30ebff74a0d04d32e1779cd89c9cc286956e22","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"337156a75835c2c36f0a1c4564d22ea4e07ca62fec4c6a8b236b2410cf0ab99a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"d671e0f2105d61a4b234137bc4037a5a47e1075b492c8cef1c53fc5c7e90c7b2","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"3963a5944c70f9871da5f9360ebaa8cca7acad5723ed704b9dc324470aa385ae","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"aab39976b297f952700aa4c3c6b9334f465e0ef5d5e9b7f5487dc85309ac2c5e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"82f5829a26f52d5f915cc9b6f3d4d5f46822339e64be3c1bf181f721fc9aa76b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"872b67207345c4d336a688444329f119e10e7179ee768337661cfea097fe6859","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"8ec48b9721eadc57d701fe16f3af931a68908b4f41b0d51443c5d35191a28dd0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"6fe1d35bed9013ad483d2c4324d7e04ca3980b83905531885260f52d3ff9d866","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"d72251527551030e47f7294e2ad798c0f13b07569037d8e616e85bab7d45149a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"3593b42713951408021d9c61a412fd23c864ca68b769b7f85afe9c8cf7cd7dfa","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"09bb1ff40b0569947279674af56dcc67e7613243de4c62c1bbfd1f774f87d9f5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"1504de2aee8013d9d8de98beeed95c97d0b575e510a7d586762a98f34b2624bf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"e290c7fad8951618168c217bafdada93a2826496b80f14fe46ebaada7ab91c00","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"7d2ca854b37cd638e00ea3154139ad057a5a577a59665b766a474eb8c8419924","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"5f1361c856777c453ee6f49bf88ddf7dceea57d74f1c2af0699f640568eed80f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"2f7d8cdf9f4849c62680b0299c78481f4f53a4a8e2f1364f6a8a06d645c2b232","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"a956deca18493eb6d990267411ee2889e244e2f70e070d214828d4c7fc2c2a45","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"84e349310f4cd21f9105de4de9288af61702756f603faa14edd0a0ad819dc869","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"6fc467e0312680e9000c81bbab8addd34b04be558598c2f7c3d636201a5b7722","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"ca2d0464831f1691f1dca08382049a17fd5b322005b76cf13a3f76cca46223dc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"23942e6c81b1f3fc6672de43d912cac3543b43ad21314e6f2c4484d7577b6daf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"0d2f0023860ea739cf5caa70ab52cf4b086b60842f8d4cf180dbd261b2dbddb7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"367dbaa627aa1da7014db85a404cadff8e5a85ca3f9440b43c3c407a1df3194f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"3f2c02eecde0cae5e5c7b3c97698f1e4bff464f2050a657807970025b7ee28c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"e44607361df7d3f8b51a8bce643c2eb27b4d892ae991d6416db6972ad6c5c034","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"d6850c4302534ced88625b67c6001a185427a9798b88b8745e03db1bc512e3aa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"ed1beba7747087db91a6cb6585e956a90e0fe11c8d7e2a0a13e1b18db2e47b71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"2506a64f3a3abf73e3c07c80f1ef3510076ee793f6d282320bc40a5370e6dfe9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"062b2e1429f42ab9065c5486b4b2caef445ca5008fb3a0b50d96692af6286e6e","bytes":5340}},"requestSha256":"1977bde8b5786372cdec9769bfa6a6c3255cb5557f5890bdb3b12119fb575c32","metricsSha256":"c0fc73676fe083bc5b51e86162f58bd541915e52f4fb8691e7b4897b08a967ff","continuationSha256":"19b204869245eb4396c775334213d23cbffb4b1935d11a2e3acba5acd97d7f80","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"ca2d0464831f1691f1dca08382049a17fd5b322005b76cf13a3f76cca46223dc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"23942e6c81b1f3fc6672de43d912cac3543b43ad21314e6f2c4484d7577b6daf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"0d2f0023860ea739cf5caa70ab52cf4b086b60842f8d4cf180dbd261b2dbddb7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"367dbaa627aa1da7014db85a404cadff8e5a85ca3f9440b43c3c407a1df3194f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"3f2c02eecde0cae5e5c7b3c97698f1e4bff464f2050a657807970025b7ee28c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"e44607361df7d3f8b51a8bce643c2eb27b4d892ae991d6416db6972ad6c5c034","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"d6850c4302534ced88625b67c6001a185427a9798b88b8745e03db1bc512e3aa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"ed1beba7747087db91a6cb6585e956a90e0fe11c8d7e2a0a13e1b18db2e47b71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"2506a64f3a3abf73e3c07c80f1ef3510076ee793f6d282320bc40a5370e6dfe9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","sha256":"3ccadfb95cdcee4d3361fe466f9eb129a3620d608fc58149928a6d03de973d45","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f8e540bd529d78a9aa98e1a098e25093fa316ae3e501786e893c9fd29511d3a2","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a5ac5094a6c0730d756cc681f27e85e69943bb37ad7bec445a8e2cdcd6c60859","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s44","horizon":40,"attempt":1,"job_id":"2480640","recipe_identity":"06cab3017878f770defb07f5660f005af2725ecaac12d93f5119e139f2de5fe2","loss":3.529488725798317,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"832e40317a184ac7ffb32f19854708ad4a5ea17f2a16dd15075c965cbaab243f","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"401b13f939927a3948c6e82fe901c7264e6ded02489b0864cd39efa35c8906b0","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"3fc097c392682f42027086a2f10b6f8330f9b71a633e9fff77e20bf2ae7e205f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"653466deea260bcb2f9c1a1baea9c0986b83cc1da5c273a80666d590cad50074","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"2d7103ed7a729869a36ddbc3818e6195ad74d495ebe843cefda1a6790d5c11a2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"e02d6d65d15c22cddfb32dde0703a2bbb8f8509a91e1f23e6fdc077a5c3d8f6c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"83267ff078cfeb1087f194e2cd9957a305626d41857771b95e115320f23369ef","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"60d3ba1bf5aade3298378f48587820925e6c90cafc9ac1f703ab43a793c6d0cf","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"c93a51e57c9fb3a32567f04c8c97142339e0e8731df5bd2089f1f8613c29ee4f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"ba0b17a519cf99fd25b610a9300a5f40a8e3380b59d59c782bf03d66ea189a03","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"0d35c5dd8ea339370b7f05d93219204bd066a1719c3085ee5c6a0336177907d9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"39b1c721667169f932ba73630e17f660fa1edf896f5fcfaa80a71c5369c82e02","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"546bc77a7fefe226b815b76bd70cda86ae70089ae7e5d58dc251ffdff1cf35b7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"34c1012be3f1813ea020716b231c02c270f56f8c1774bf86430403b3ca4e54e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"0e96f3c56a8486adeededdc2a6e9dd19ea2061fec03f7744401e9db0fc73f6d9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"138fcd080196e839ba2c335e0600f0d072281e76b4265a6ec2508b5fdd7a64cf","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"f35eb95f2005913f69856b2e3a3c89553295884f53da8633756836aee240f5fb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"74332d4588d0e8ca928152bb633b1fa5b18ee223bf7c581a0ab6e7a4e452c52f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"48b62dae8acc162e29552b3d8ad67675a01f372dfda548c78d29b981524e0725","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"276e29160b3c6fd87617396d409925880b948aebd316da08e9970e191b3dc342","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"99a02485c41d577ca88dbc89136ad136a6f150cd28203a986c3f9f72b6efc989","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"8894bd126cb78345483c19ee2e7e0809f55e6400a1e380761ddabe8365bd8ccd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"2c8f1e6e9bf4367d65e5025bf6969e9566ed269e35fdb1750ff20f9cd90b7ba1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"3a2399ef2b18e11651119fe16214b48b3ff724b46e9f2957c00b667f42c2160a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"f2d8818cdd953ae4c167730dadfa456e4cd5cbc2061bd1c1b5387019204987a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"9df57a395702404dd15634d3837ad7a240eab0ef11d756c3f73cd63bcc94ee0c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"1ed6b39d2faaf4753a53cd2c664dee47552eecb87d9326280374aa54b92619c8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"50c22d8f702e3d8d0fcd76e48a872386c7500c6817a2681ed827a56f8d6769e5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"eaf1e993b169912cee3ba381aea3e183437e97b9fd9d135da1d9df499cd62119","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"ad41c6afdc06d1ec1cf81bc0f3a9695370d5eba03411a2a37869281db3a31baf","bytes":5255}},"requestSha256":"5a3ecad73ae5bc464b03102b583162611de414913fad6d62c4120721501943a4","metricsSha256":"5d33c17e04ec7a39e206824c29ec09e8243e4110ad6cb81f32ecf01512d24374","continuationSha256":"cdaa964d344e668153870136d557e3ed5ab2d9464647e2177fca010e7596f72b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"99a02485c41d577ca88dbc89136ad136a6f150cd28203a986c3f9f72b6efc989","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"8894bd126cb78345483c19ee2e7e0809f55e6400a1e380761ddabe8365bd8ccd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"2c8f1e6e9bf4367d65e5025bf6969e9566ed269e35fdb1750ff20f9cd90b7ba1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"3a2399ef2b18e11651119fe16214b48b3ff724b46e9f2957c00b667f42c2160a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"f2d8818cdd953ae4c167730dadfa456e4cd5cbc2061bd1c1b5387019204987a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"9df57a395702404dd15634d3837ad7a240eab0ef11d756c3f73cd63bcc94ee0c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"1ed6b39d2faaf4753a53cd2c664dee47552eecb87d9326280374aa54b92619c8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"50c22d8f702e3d8d0fcd76e48a872386c7500c6817a2681ed827a56f8d6769e5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"eaf1e993b169912cee3ba381aea3e183437e97b9fd9d135da1d9df499cd62119","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","sha256":"bea1c7dd0b61b8e0ea544a6495d44a66df2cc4a0be1da14266641773e29a5b61","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9038d74f3bda179852dbddbd7b19e2b1dbf3baf8cdb77c401a89ec202f26a8e9","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"06cab3017878f770defb07f5660f005af2725ecaac12d93f5119e139f2de5fe2","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2480641","recipe_identity":"40cd3959b6054bc530c7b0da8f9bf2c50c3944d4667e0bcd1f78d65fce15dde2","loss":3.539456550603085,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"3ed00446b3d73bed08317c68e9afe27e36a1d0daf09bbdf78b86cbfbf8571b3a","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"c21df92a975b5fb7763cc8391efd19ab09648985a9389137b236545ebaf49940","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"35a310935e7b92555892ce41b877867f7924dbfbd64f17e4c22f1ab85c5064e3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"eb7f8938ddbf586f62ed2147a2fbe6a7e30aaf24b75dd4f87f85a3900facbda0","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"565702ad2352a32d2c05090a665ba6e2871169f970584204390da51a81f968d8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"6a3c36602128b327f72a0587b10b169d83626790547612f6a748037214e8b407","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"d94aaf97c1fba9a17e7c10175f78cabbf7e8d27e8a4e61cb8ac2cdc123b4a592","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"68bb9263666615cc1467845520fa9918cc9e79e508343d9cc13aa12e231f79ae","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"f70bcafcd4ed6b8e6fc53a51e4a9204c57c6c071911d2dcc5ce345b9351c169e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"76730afc000d2646e190d50b475444fe80c0ed3446c0e686ed459e79691c3f11","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"8c1c3a3505f17ef76027b2866abb3562cbdaff8127525248723f94de2895619c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"7a3f326294b3e205eeabf5a74a543793a797490719845a61795ac9c08820be94","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"2b4b6bc0aafb88c5ea034b39940ff33c545e03387276d64ae130e45b116e82c9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"f5dd7d3d7127ffdb8b46d7154053a0cbcdb26c9f19111f3a42318eb7f3bbdf9f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"360e807af40e7aca9e7172a313d3702e7e6ce1e17aae82fc7a2914500d387152","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"dad53b8dbfed92ef06dee7cb87c237533d255a68d798640691c172be50785803","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"b6c9efd88e074bc15a9e9b4bedc874a4708a6581f8eba88786fc869813f95f5a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"fdbf1b78fc439a5c0c89b87a3523e32a5812c9ba0731acbede971155af29bc0b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"13cc657dae9bc3c6a513984e382f3fcfed76890f74f57a5b01c08f551a464add","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"961229ad1587dff0c5c2d242e516950d3bec216ed43f2c2cafa59044556f7172","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"c79a8279e9b7cf8472affc37562c718a514abf7792f9c0e821117ab378658958","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"657a38e519b132316b96014559c2421fbd84658461847b595cfac042b4a0f9b2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"659f2ad8036fd3aa5cdd4dd06eff8259a61c7cd89384c3f339280b57fe3ff91b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"6e2a7bfcb3c67ff62448322e338393dd06db2948c170e00d45dc33ca7a1786da","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"181db2a8546d09fa5cda5c8559d216c9557280d903cf9b802ede5560edb4fa24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"f45163e3876c72d178ae41349634c5df9bb04a2f11aeb63bc8db2b3db55ccbf1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"28fb6ff86fd81df4c60a2bf3bf163cd4cbc864d839e0a73cad5a7ee360e5f054","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"fbf829a366b8e3ec211ce2ec349bf048e3a4fc03806e444725132c9d4f89706f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"31ea69ee629d639cc6d750190412f477287600da9d7de44bf9fd6a6a149f39b3","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"1c40cfaefa772b3ad16fb92865299aa7b9f4c335a9e5885c1882180cba4ec953","bytes":5341}},"requestSha256":"c2333d3e35c066c344379d4d84f9c41261f9a2f18e8c050f815213c5e040166f","metricsSha256":"923096af6443d9dc54bedb54499db1fc7a86086d6c2165c9fe02e735a62cd9fd","continuationSha256":"d8f6d469616d92d3d0ab1e736046fd90a286ac47bed772468eefa36688010fe4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"c79a8279e9b7cf8472affc37562c718a514abf7792f9c0e821117ab378658958","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"657a38e519b132316b96014559c2421fbd84658461847b595cfac042b4a0f9b2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"659f2ad8036fd3aa5cdd4dd06eff8259a61c7cd89384c3f339280b57fe3ff91b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"6e2a7bfcb3c67ff62448322e338393dd06db2948c170e00d45dc33ca7a1786da","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"181db2a8546d09fa5cda5c8559d216c9557280d903cf9b802ede5560edb4fa24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"f45163e3876c72d178ae41349634c5df9bb04a2f11aeb63bc8db2b3db55ccbf1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"28fb6ff86fd81df4c60a2bf3bf163cd4cbc864d839e0a73cad5a7ee360e5f054","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"fbf829a366b8e3ec211ce2ec349bf048e3a4fc03806e444725132c9d4f89706f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"31ea69ee629d639cc6d750190412f477287600da9d7de44bf9fd6a6a149f39b3","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"98a449e22c2d307810696223597c12bad397fdc23595e1483f059c62650d2390","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e49e7bce358bbb6d5a2c3bface61567b683447f5f5327383ae2d294544e4d9a3","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"40cd3959b6054bc530c7b0da8f9bf2c50c3944d4667e0bcd1f78d65fce15dde2","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2480642","recipe_identity":"b7d05813332bd3605d4660ea6c2061638407af8f52f2a9a69a56eeedf1480bc6","loss":3.544250028278091,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"4c2419d793af820c92574fd28e5aa085d5defd5ca109c14233311d6bc0c5ee16","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"e3e1c9783a2053868e8260cb91c7f3404d9f27a4aa443155438ceaad085c8d9e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"fbf5d9bca45f849944b2caed4e8dc5281bff15d1b8f1f89b3ca4ad8e16591bd9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"783865e1eff5b8d4850c936ce8849b94be0f3da91d60c0cc998fcd895ee7e65a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"d94ae55dd09e8bb46265971ca7d7a304c8ac79983b00cd18177085bd80c8a6bb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"0106d90f453d823dfa3ad2d97b18fa2dd388c46db4ac954f5b4b9e7d15c39d2e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"291b6f6c9b5cde7007d426a99cb3d40da4ab2f82453446698aad7e5f99af0bdb","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"0d4d73c46abe90ff5e1eda622c094d81046f662d928f6cbf3d1136e7fecc1039","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"25c4e619a4cf6977b7c922574c2b4944da679f56783c4a9a2784ede9f09a8149","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"faf01bce2d100a38ec68ba2067e759c7f00cb147c439278fb7cac060f3483661","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"ae1b6c4d2cbb85bb227ae2458c822cccc9728f8a5a7a04d92b2f433c454233cb","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"43a645ca3a9d1cbf379fe302eb0752c12c7eb105527b30b58ab83e2386dfa647","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"58b07a82aef9b3133a892325cc6082763a99fc21982cb61f36c3483c86a67792","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"0ab182eb6c1d1d4f903de8febba184de574a70f62318db9939023a438a4c146f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"3a2173b1d4f6809e8fe6407cf6336c2c9b93fda0de1d071c8af97270570ba4e4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"6322343dc02eb45a4910f155a6077d38f3112fbb4891659b77abd2f411820392","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"8d4c66d58c8b7877fce6eabd9a657e6e7b7ab82402159c2b652e06ef1c92d828","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"aa7459e2b5934c09ef59eeb535cbf6b4e69419e5cb09559320bf4edc6bfa5ede","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"ec4aca1d6d9e5eb66863c0f12a0f740f4eb25e4270e5be98cb96055786279f16","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"4eef365842d8941d25085b6637d80fa9bdabcf016a95bb657c384dfb49bf8c04","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"c16c3327eef088a36b775431f53ad54e114fe5a38b3bd3eb634f932d8c969812","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"3b4d6f9929c1996dabe9471f4c4905ca30e87d872d0089b98370659218236b80","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"6529d1668b467fa580c7962a7c699ec9cf516ca290f6fb229c84ef89acbfd7ea","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"8bab287813aa25d8c322eb92beafa1d1306fa32eb7d0b50722915001d1197a26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"503f7ee6cf3f67ef106751984bb43be8f4d430ece800926759acd7c7071a8a74","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"0b989c3ee62a1f65b8319855f2b72dd679e89122cd04d34776933bfcbe0c6161","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"6b402bbe43871acd9f7c70718c35ebb50a081f1d79773cd3e6738f93dfe2f0d3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"32bff50e4c52ba0395be8c4f6a7ec9553d175f352afb27b5d938d0ab17dc3b5a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"de9003901a19099e5af08b0bf83f9e00f4953e4b85fba790a5b45dd65d0dcd9c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"271a785f60acaff205f76cf5c4aa280d2458fd041b63bbb96e86bce9a429567f","bytes":5255}},"requestSha256":"d70074dba291abaed7a7a2608de04a0eea846443c1fd921c7e2233a81161dc39","metricsSha256":"462c085a16ab8747a2cb15e3627a0595c14d37842448ab25c31dc5f1c33cf30a","continuationSha256":"2a3c7de8d0a08f6efd3a3f76b47435de606c0c6d0cc6934b31bc48f99db99b8f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"c16c3327eef088a36b775431f53ad54e114fe5a38b3bd3eb634f932d8c969812","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"3b4d6f9929c1996dabe9471f4c4905ca30e87d872d0089b98370659218236b80","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"6529d1668b467fa580c7962a7c699ec9cf516ca290f6fb229c84ef89acbfd7ea","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"8bab287813aa25d8c322eb92beafa1d1306fa32eb7d0b50722915001d1197a26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"503f7ee6cf3f67ef106751984bb43be8f4d430ece800926759acd7c7071a8a74","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"0b989c3ee62a1f65b8319855f2b72dd679e89122cd04d34776933bfcbe0c6161","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"6b402bbe43871acd9f7c70718c35ebb50a081f1d79773cd3e6738f93dfe2f0d3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"32bff50e4c52ba0395be8c4f6a7ec9553d175f352afb27b5d938d0ab17dc3b5a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"de9003901a19099e5af08b0bf83f9e00f4953e4b85fba790a5b45dd65d0dcd9c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"f36c50387e47c890ca6ebe08dea87081d3210b1e41e496a5ffa76a2e624796c9","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"41c3151fe8f5257a4a7760ea42c8e5ef5bd39fd081f9465c7211acf69ea8c54f","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b7d05813332bd3605d4660ea6c2061638407af8f52f2a9a69a56eeedf1480bc6","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2480643","recipe_identity":"b9364d9d515758793624c2b37e80ac05d11c84d377a74d28d30e976dff013e0c","loss":3.5346538530127796,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"ddbe486444ebb78a65ce3fe7c3dac9f887ce41eccbc7495dd224e523f0353344","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"80e61f6df36bc979a6a42034a19e6a7b8233a77c9e065cc57901d1b27fa0b83e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"c759823f1ea2aa9c3b9a46bdae4a0cff5e14e0e1f5b52b37e2f82339f8848978","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"f1339f31b6a5fc01ed4e84f6b0b0e0380e494a479894bd8be6519124c1fda6c8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"01081fd7a969caf2c4589c2a657957347bf1e25f5692caf143155b0ddfb2b382","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"ad0a121a33d22827788bcb28a9afa5a91fb999fceb31916b4c41ea4e88ba019c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"eb12a3aacec7102cd75f0d24297ba8601a6f326dd2aea8af20903a043106c1c8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"5bee7c081083074dc6b1b15e3018da145e45f21750c2312d6e7f67a7ff4476ac","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"f51215a6d38d6c45d091f7b4d1c9766a66a33b7140b47ce316798de90528ff47","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"251ae44a2e91eb8998b464c792ecc696ffd21be8b46f9a4ee36de992a927633b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"aa51bb476372339e2965d42f3778455108f5e6f3c877229999939df6ad9b7a8e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"a1417ffb64025a6cec270931d257e33f122caa34e82cff4477ca98a090444a8e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"b8e29cb308cf42b84f169b4b45ee0a0e5b49b4e7fb989b19f2060e9d59f00346","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"403c71f6d85b1392c8f7dc4a180eb3579bfb22c94b1ba3d4f491b0edf7c49eda","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"e9ec8cec0196d00c4664d93f12aba19411b86a6dfd403c2146525915e5c4ce3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"7ef696ec148497c4eace7f900fec06fb6a1569cb8ce207fc9c5a0246007f9724","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"3f42f9772cb11b1ab60956c524f36b8b9e274f70825581972eda9854a1fbc84a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"379ad78a07d0aec248bb63b1129c1538cf62db935d0c1922d1310950b6ac0acd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"35c96897e6778111cbb37bd3174b39951e301d2be22079341254520dab73691d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"a6af5de1b31efef82c684fb95c441badcfc8ca5fbb4b4bfa83aefaef31ff8d43","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"19615c27af98a18abd11b4d4f509593a9b703199368c6064d02988885027ac22","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"3cdb0feb4de85f837d30d3558269d0114a458ca15973373432ba48a086de60ad","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"453c0a895e05964a859922d2ac754b8d3f7ddefc207fe56285269facaf631e89","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"e4f079818866b3c3e346e9c62403bae9494753ee75497efd143597271a44ada8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"22018ce85a816aca5ee7197ca0db1b592e2bf4284b133d3a4a7a07e15efe77e4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"bd2d28ddf243e8f164db5c34e19a961499ee9de1eb5d4f5c63d4d313684fe0da","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"f5fec8c578765696dd6ee71ecbbd13fc8f093189debc510f067a3251bfa146f0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"c084a00898711ad4a21866fef48f604fa84cd4dd0ba4e91cd01c512008436e68","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"22e5ad642dc89a6b8b26d566be431a6dc93acddbaf8443fa7425362021249d90","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"2c6aeb6eeabbf1472538796983a7a9ea5e698ad8fff926cc75d072b6f4077401","bytes":5312}},"requestSha256":"3a34dc843b5755da9eecebbf47490d88bb31b8b10caed5661ebe44f4c9fadefd","metricsSha256":"437e1530cec70d13bc36eca93aff565165256d62650de9970a84703aff05cac1","continuationSha256":"a5abee810425cc5c265b4ff9bfd005659d401c1c17e5adeec03094ffb129a5b2","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"19615c27af98a18abd11b4d4f509593a9b703199368c6064d02988885027ac22","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"3cdb0feb4de85f837d30d3558269d0114a458ca15973373432ba48a086de60ad","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"453c0a895e05964a859922d2ac754b8d3f7ddefc207fe56285269facaf631e89","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"e4f079818866b3c3e346e9c62403bae9494753ee75497efd143597271a44ada8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"22018ce85a816aca5ee7197ca0db1b592e2bf4284b133d3a4a7a07e15efe77e4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"bd2d28ddf243e8f164db5c34e19a961499ee9de1eb5d4f5c63d4d313684fe0da","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"f5fec8c578765696dd6ee71ecbbd13fc8f093189debc510f067a3251bfa146f0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"c084a00898711ad4a21866fef48f604fa84cd4dd0ba4e91cd01c512008436e68","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"22e5ad642dc89a6b8b26d566be431a6dc93acddbaf8443fa7425362021249d90","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"2631169a9216983fdf942ca00e89ec857bcf2b80df29846ee2ce674a4fb5378f","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"75e0318012134a782cf50c0e2b8345ff6ce93fd2c5f2131b614bbae2bdc872f1","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b9364d9d515758793624c2b37e80ac05d11c84d377a74d28d30e976dff013e0c","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2480644","recipe_identity":"fc05ae5daacc786e30849162ed387493f61093acacf9a2d241e0813f1d1711d0","loss":3.554913063305192,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"ee2c1e538f3339defa2d8ee50639116e8695d3e9abacd014459da0a36fa07b9a","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"7d98c9ec51c3c28e85349746818243fb85e3b40cf156d5e097fcf33b9f0e30b3","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"193fd53b1a1e11307ed84c285899a71040d2046629f3b46f206619534b00fa5c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"8ddedba124ddf7268a4f88564989f262f838520a0ea3a8a699ab27c8e59212bf","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"9530daaca049fadd7d28d30f45b8bece80d958a2e394a2d812044043a0958caa","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"15585202690bf87edcb1c3275eeb03e009120c5dd18a3f7e974a9ef7980d3e40","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"d11c6c3e4b3bd1d5bd4e35437a452fadd8f3c8741acd64e9cd863d48d24d4e2e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"4b32cd7b7db57aeb841b2d58b990c034c083c8f0f6bb040e001e0d47afb965c3","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"5ad7e6b6223133e230b0ea5733ffc945bf9ff34ee8c4a3419a76b3b1e09c281e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"9a4085bb374edd09824d34424dcb162f78ac4ec962ae2640ac27d8a5c52fa70e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"35b8dd732d2bf5a71d06c6e25509942e9045c02dd987eee1072669910ea39509","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"020ddc111194468526875c14a198815803b8c8bd82d2924fa6238658988c0e90","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"228b1bd552a32d9fdc132496d95ab6287148f1a483487f64b13f8b472d0ed5b5","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"c760f067ad61abe20b2260f4ac3b8a7526652611f1a39cea25a4419d63112d77","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"16d313e273bd790138851204cf97fbe39258d303d8fd32996544f879e65d7511","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"6fdc953cc2c55e2cf30f67daeff1f89176d1196a20af08bc24de33249279073f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"bf2180b1ef970e2506a53b1b77220a6dc5b43106f3b48dae18aece583f14ab16","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"86954835ad428fc5ecb500f105a6592bb773b68e65d239f0bebceac53c843519","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"46e5c0ee3b60f39bce2dde7892a391d30fac807fc264d4e46724bd73b218884c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"ac1ccd18e3d5fed1cfba75dd1b0c30358cceb9e3f6a164df6ffa3dcdc44454c4","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"0cd6460a3e1b9fac6e2965eeedbe2713bfa4be6562fe3023554936e484cd4ca9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"5817fb136ef8f479704f9f6f85ff12dc5e7a9d92bc673f7cb05a03e38ba57500","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"465bfeb8fff892dcdacf05fa2696e98db9c1e88e052c26110ba5c8600ae5ae6e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"76fcffa2116fdb76a8daa16d5cbfce40efb8312eba88b15b4562534640f71cf4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"872112b5044b78986c9921108eb234f415ad703e89ff85bfe3e61c93b01853f2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"0c543e0441fd52930b0fce4583d482cca364de5c881fc8e43e968e84f2d3797c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"d8561eb37ab17e3eabb5a495ffed77e30e3a727e3333ddeaa0ec9ae681685a8c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"a5b72169e39a164bf96f821d219448454ee074995e74c3f9c400d314604283f1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"0d47725e472035505a43b7292cc677b61fc19e2a62ef67881f76fd754f82a4b3","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"02cf83b0e98ef4f9b51bea348be23a144ffa4bb4709d4d1d7c2043aa7334309c","bytes":5312}},"requestSha256":"fdd2530992a1cb63fd9a34f003b29c629865597d447a82d7c7c0fad4fde43b58","metricsSha256":"c6fd45db568c3fbe06eed6b371b4e89b98a186b5cf72f63b3fb6c31f115ad475","continuationSha256":"0bef4e1299c118d9b1c852fe7cf8bd22ac506fa8c72cb524d628186889d408fa","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"0cd6460a3e1b9fac6e2965eeedbe2713bfa4be6562fe3023554936e484cd4ca9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"5817fb136ef8f479704f9f6f85ff12dc5e7a9d92bc673f7cb05a03e38ba57500","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"465bfeb8fff892dcdacf05fa2696e98db9c1e88e052c26110ba5c8600ae5ae6e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"76fcffa2116fdb76a8daa16d5cbfce40efb8312eba88b15b4562534640f71cf4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"872112b5044b78986c9921108eb234f415ad703e89ff85bfe3e61c93b01853f2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"0c543e0441fd52930b0fce4583d482cca364de5c881fc8e43e968e84f2d3797c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"d8561eb37ab17e3eabb5a495ffed77e30e3a727e3333ddeaa0ec9ae681685a8c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"a5b72169e39a164bf96f821d219448454ee074995e74c3f9c400d314604283f1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"0d47725e472035505a43b7292cc677b61fc19e2a62ef67881f76fd754f82a4b3","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"d7f329edf407ae7b90b83ec51762061630a420969c9885acd81719773cc0d570","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c943918421b102d2e481525cb4c834f6cca701e138b10341dc89a1ef286c574c","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fc05ae5daacc786e30849162ed387493f61093acacf9a2d241e0813f1d1711d0","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2480645","recipe_identity":"4e5e636520f55be752973fcad4fc6a8567a9f2b5993a6c550323f9a5d6cb506f","loss":3.5514979037753567,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"c0a1f5069ce26feb40ab5a4e8f27e63abdaff7133d11bf2dbb3c3c7014c04e2f","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"bc9e5010f6b3760805f74156c2b0e7332767177a8b4968d8ecd4eff0745ff6f7","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"8df7bb367c6ed50ab17c6bdb525d95711aee6dd2cefa890ed7bb6bcd633150ea","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"19278c4dc62252d87ada0907535c147a8a5eb6fef4be7d39b0c22185140f4746","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"018afd0b069edf8897ea67f35d182b114adc4dd41660222a563c3c29dc409af8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"1a5f6584468289d36fb6b25db373ff03f18dc9cb6364d57cc146aea254c1efef","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"e544e9d444cfd4f19ba13df06a2808b3fa7722d994f7af3de0b0bd7ca5a4e9a6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"12316336524149cec40bca8076da03ee72490c34b5811220ab7aeb9bece71858","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"6a9fbb3831e67fcfe2f03c3848362f32777b3992ec3feeade949077b6f5632b8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"12a3d1246b43e62fca6dec59d7db7299b5545ca50eed8fa83a5a7fe979fccc35","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"20e8306eb27c5a66c4332aa5baac4d99f3aeae2818acf1079cac2bba1e4316d1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"144dd5dbfcfc29acb45672f8c2906f2c4ad2552c31d76f394943a5ec34122ee5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"f90c194ccc3c345ce203c7ca111114f9b8d8649941d15fd11469156babfe6963","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"0985fcd9e0068bf27bc20194f465ebe1e6afefa471047d9e8af9116241d4ed52","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"188eeeed97b77d111019ee5ce3b87782156a9dfd205e0cc4fa77d8f998df7e2c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"343cdc92aa2e439033512e3d8957fbe4d6a821b302d0ad5432c5167041c80f7b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"2c95cadddfa54b2538cab6f3518df768108707513380e323e26fa673b663b747","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"b868414fd39e7d067204dbe953f6dd1d7883ef7fa5fed7c929e090f221224aae","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"85c5f015cfba08bd6cded1a0cbdd81f071438538e12087cf129b02f4836449b3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"bd4abe78e7439d6e1d64a9c602f89234a07049cf1b178a65f6237bfd1fdda5d5","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"60551d8921649c2d4153c886016f003307b4f26ee41bc73752b9c4e5859ef2c6","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"1c375b36852d0784d3c9b34306f7b93231455f77b8e041e29114eb02a0b0a97c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"630fdbfc3fc99a54cee2bfecef2a12a837d7f31fe3a4b38bbc3e3be247af6a92","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"843ab2b4a4b1f8cd9bf89b643577eed30d1ea742e639cdef5bc0d46189a1f677","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"ce45e5441aa785d06891330726cd2e062cfde7493a95ddf6734c9a23b8351dca","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"c4cd2c8b8ea4fb2974486774a2f87182739ce6abbc40ae0aa227b369cf017f77","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"75b09b266c03179e0e090cc1566ec0a88052b894ea7bfe11e792798e6c6f2b7b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"c4492d683260eda7c83d98bba903e718f98b267c5ba582e3aa7816fa91962c98","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"8ad1d9c1bb3c536597a32627e514c5a5dc5b286e5bdb4697e13ff38d9a04f748","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"619a3447608f85064251428201ed670e71ec59d1746255e4e5a2cf95e260e9ba","bytes":5341}},"requestSha256":"73921b2b6952e150200ec34dcc6f5ec71ae91756aba7900db74f07fd2f5891fb","metricsSha256":"de52255556a88c89f7636e862f25bcf97ebb113ec40d4f68dc5d9fcffab9ddc6","continuationSha256":"32850d36c3f19faf6ee693e3663fd3ddd407917a6c7a9929b2379f345d50be2a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"60551d8921649c2d4153c886016f003307b4f26ee41bc73752b9c4e5859ef2c6","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"1c375b36852d0784d3c9b34306f7b93231455f77b8e041e29114eb02a0b0a97c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"630fdbfc3fc99a54cee2bfecef2a12a837d7f31fe3a4b38bbc3e3be247af6a92","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"843ab2b4a4b1f8cd9bf89b643577eed30d1ea742e639cdef5bc0d46189a1f677","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"ce45e5441aa785d06891330726cd2e062cfde7493a95ddf6734c9a23b8351dca","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"c4cd2c8b8ea4fb2974486774a2f87182739ce6abbc40ae0aa227b369cf017f77","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"75b09b266c03179e0e090cc1566ec0a88052b894ea7bfe11e792798e6c6f2b7b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"c4492d683260eda7c83d98bba903e718f98b267c5ba582e3aa7816fa91962c98","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"8ad1d9c1bb3c536597a32627e514c5a5dc5b286e5bdb4697e13ff38d9a04f748","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"f0aebdf12d3213734f79602571b552313ad259ea606ac1a903e7909701be519a","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"89613778618199ab0e5b8bc0c6530a469115347253d7ba45b58deb8fe40fc7dd","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4e5e636520f55be752973fcad4fc6a8567a9f2b5993a6c550323f9a5d6cb506f","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2480646","recipe_identity":"e534528ed3fc9cf6a874bcff9ee18f430ad364aebf21828163f5e5085f8fb619","loss":3.544381138406371,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"86a8e101ec61e67d65d33017e1ff9af931b57adc552c63da4ca2710dbcf3d961","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"f80de8d12dcbb1a3de35a32f8b41b02cc89014bc7691de8f94655f2ccec45bab","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"4eea0fc185418c4f4ad1b0676d605ba4a25f39114473bd366808e30690c839f8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"b1937d532019c3132112c335b69859c9a19e66ee0c903dda6bf993daf39f4c9f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"ce20d55ebd339264415cd5d22d74391143652ba715c8080a7d948050843ca636","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"1b91c6a80e0937481a38d4282759da51f58789ec93e390223a0fb05061f3edc7","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"a4414d1523e67cf280c5125b411e2084624f32f85c0298e560e60d3785328722","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"f8e92633bf11e48e4d1a9245595082da10d78793b796e066ec40b7b6141f5842","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"c7d869ed52d2fb62de8af4dc3a9783d6e35140624fbbd3ee994de5b06c31cded","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"2c29522c9721d7e00e082ce4928d862438b8fc34bf9b75704e0db1995ebdfe64","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"f4887562add2e736cafcdb941b5a3823c0fa48d45102fc01d60b93e298396d04","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e1c4ae39e576aa3746be804403ca02aa5cb8252055ef489fc4d97c52c22bbfa2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"b2fd00f41c28103e718df0551baf7c4ef1e5713b70ec2addfd7a40f8b9c41db0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"b09a478e8d88c8a070d509d7312024d8619c1d067f2c6cd0fe2cef6ff1fb2733","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"fadef1a94be62002bc04e4cfe77cbe87aeb4b6403bd78338bf01604c1eea51a5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"52a4fb5fddfd0b85de684d26218c68218dfa92c544a2ee6bf0988a02747ec490","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"8d8267e87d0b9fcfe8e24be7a9141cc618af60c8bfd8cbd61d73500bb66e337a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"4e9d2999627a2c54705cd3fc4c415861de26a4dc547540a7a59f5377eb735be2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"f44ee2b19a9b37b5aa9becd8712b8597c6328091a76f14a6e4747697324f54c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"3b4efe24b1cc28b9e58f252c52641ce21938ec25231a2d5feaafb6c98f49489d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"fb2b141de2ea51be072d55af786341becb093df2f0dac6e25bc50c82c7fd988e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"07528ea65f5e7f48d10f4b8171434b24f544274208ddff82bb6926ec267ec979","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"c4c28654ae661042813d8d6a179a037ed111f756afbbe99893ec7407a9ea43dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"24e0ee822c10f725613b6481d7c825e7418874c7b425119d2a87503c136bea2a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"d29de7d0f5c0a0c71cb1a37660919b5e2533fc18a27e2ad6a377205f0ff1b9da","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"e95633ca94693e3e7a5a1c4f9b7f43fdd22dd3db31ca159ce07a704d3f37f457","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"19797c4d01d02bace6483ce0fece53efc9c6a43b45da8a12f495d7b3d4eb6e96","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a7a7473a82c9f0154e5a24740fea765f57259a9d2693e6bc8a0cad72b50c044d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"0b11ef69dd4d2269355b3f454cea991fa8a1bd35dfd703ecbf83069a1f2448e2","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"08bf1c2b76c675c8cf6d0013b3fb95a8923c91e60447c5df668feae4e9082cff","bytes":5253}},"requestSha256":"69c6128be566f4481d297c029393a4e252a32226c11170030e87dd69cca67f84","metricsSha256":"b9f03361e041a4f0bfd27038b3822ce8875761641540afcd7e163e6cb48ca99d","continuationSha256":"3e196ee1107bfe341718e990eba0002ad047b81ef55d1594011e3e08f72c9e58","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"fb2b141de2ea51be072d55af786341becb093df2f0dac6e25bc50c82c7fd988e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"07528ea65f5e7f48d10f4b8171434b24f544274208ddff82bb6926ec267ec979","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"c4c28654ae661042813d8d6a179a037ed111f756afbbe99893ec7407a9ea43dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"24e0ee822c10f725613b6481d7c825e7418874c7b425119d2a87503c136bea2a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"d29de7d0f5c0a0c71cb1a37660919b5e2533fc18a27e2ad6a377205f0ff1b9da","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"e95633ca94693e3e7a5a1c4f9b7f43fdd22dd3db31ca159ce07a704d3f37f457","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"19797c4d01d02bace6483ce0fece53efc9c6a43b45da8a12f495d7b3d4eb6e96","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a7a7473a82c9f0154e5a24740fea765f57259a9d2693e6bc8a0cad72b50c044d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"0b11ef69dd4d2269355b3f454cea991fa8a1bd35dfd703ecbf83069a1f2448e2","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"b12c71e44bf26f82187edc920550f31c4c554b40a281e7c57dd3ee1f1b5986c8","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d38853ed86b24cebcfbe30f227a1c3bb504a61f4d08c00883b7c0aaa9a77db6a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e534528ed3fc9cf6a874bcff9ee18f430ad364aebf21828163f5e5085f8fb619","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2480647","recipe_identity":"54ce58b2f5b617c15cc60e48a79fa4097dfbcc3f0580e39822985c0dacd72a0a","loss":3.575773738630764,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"83c67fda6a3da0ff15927fbef5ee8aa37e1d2c27bc623b238ec5f4dc0c3fe03d","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"e89164a5f761326da8dbd5070d45e60d04973a97d2680a2a076e708f9b724112","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"ae35895643b20bde6cf74d8009c34d0491440019956d6ffaf18249b9aac5d129","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"ae2048039e5964dc15ce903af35b16a506676d20e537aa339c57c7e42f65ca2e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"2def719a53a0c6f07cf27ccba4ff886d1d1ba1908ad337833a55bbbd15a6304e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"bba824b2e713b0ff2ef4154bd79e6b47e43c1b9ab286c455e09874366b10c57d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"4f63de56cec52010e845b3c8f343e9fbd31ba82bd64a64130d29b866fb5e743b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"15d367df2efb728cb847d52a163104c8141234e90666a5b988b2fe6e95b04a0d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"16cc7004540f8250c8898e4a24de2c4462ad6958c050081315e35df855bcf018","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"a9e87032d69cb53b38b84b73602ba9d97791a04f90dca8177050d550f95c1bb1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"755a4d1b6ba6739f0a63d2ed2d7be176a50464dd72d9bc814ab206757a6d9754","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"17df17c6a50e1e50844a62a45a009cc70cfaba31fc92a8b5881ddbeccb9b8102","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"4971b6e0fa34ad3cc35bd26fddf3f8043e93020f4ddc51602795a5ce3340f961","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"5e6ce7cd03936ad8aec7ce33a5777d8fe19e3eb63ea132d2bdda52e5cf4dc8a6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"00941a36623cdc2a4a0fc2989d8433002839774e08a5011eb3f4cfe798efbe04","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"9d28175b2178ff31a81a33547c600069eed29e992613dc690f992dfa65eca131","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"fbd27b1a22c871aadeff75aed5342a455cffedfc8894fc43a476db0afa90a993","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"172abd25c46ad700c9feb4906e28b4ae4574dd19ef8840d8c6b7f3857d126132","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"7881b0e86951497e5e3325456a4e3d701c278cb1052a12ffdd949c951c418711","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"a46a8dc5522ef8e5d389a9fccd407e34a34214ff0c4606d8fd3487f9185153bf","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"013dac99d839f2a83b8d2ba48587e4d1cec56b7efdf4e36ca0e759f5c747fed2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"3b019786c7968617025da87510943bc4f67d80c9363a0d36fd8fdfbf83f0ad35","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"ec77008d8a1d30f7ce5b12285f735b1d1c1f80cf3140c356c3a9344e6e0687fe","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"db11ad37077a81fa08872b8d022f833c89ecd1237a9f923f89cac726e7bd8d53","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"c877757bc1ac71078ed13148827269a02037a0f0940ab1d4b1ef2ddc9999374f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"51242d462dbeeb325e5e99de64afcb951d9c40a08ef2f8eca54c8f80a4df27d1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"fc28584bd1c85f0832385ea3bff6b30958d243d528207fb3013f8303e80bf31e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"bcf71a9bdc7246f55a6311240ec2321fd7859a67f6b78690b3d47eca02d8fce8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"6e982e88b39f1f159dcd1d1e285997193b48192ab628a5b2e10d40f8a82ba0af","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"6738532be31860b52d5293397992beb56054209fff96b033a7b5c1c3ec4dedaf","bytes":5310}},"requestSha256":"1daaaa2188c1f6552de64d15e29ebaf0e8f893af30a5922b69a15e4b84b96598","metricsSha256":"a18a8a70252f1a552f45720740b29a24cc3aa51a5f0e1e77e42908febdf039c0","continuationSha256":"3bbac8b42a844006cb67a8789226be8e57ad7eb44db1ecef4397052b3256de59","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"013dac99d839f2a83b8d2ba48587e4d1cec56b7efdf4e36ca0e759f5c747fed2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"3b019786c7968617025da87510943bc4f67d80c9363a0d36fd8fdfbf83f0ad35","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"ec77008d8a1d30f7ce5b12285f735b1d1c1f80cf3140c356c3a9344e6e0687fe","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"db11ad37077a81fa08872b8d022f833c89ecd1237a9f923f89cac726e7bd8d53","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"c877757bc1ac71078ed13148827269a02037a0f0940ab1d4b1ef2ddc9999374f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"51242d462dbeeb325e5e99de64afcb951d9c40a08ef2f8eca54c8f80a4df27d1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"fc28584bd1c85f0832385ea3bff6b30958d243d528207fb3013f8303e80bf31e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"bcf71a9bdc7246f55a6311240ec2321fd7859a67f6b78690b3d47eca02d8fce8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"6e982e88b39f1f159dcd1d1e285997193b48192ab628a5b2e10d40f8a82ba0af","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"bb7cb09b30fff98b82ef939ec397dd5d099d8b9e00f0793ec45c4e2261ec22a7","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ed97b3688b0b8348cc06e2f8ab2e1541f97e102336525510a940e10b3c7b0865","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"54ce58b2f5b617c15cc60e48a79fa4097dfbcc3f0580e39822985c0dacd72a0a","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2480648","recipe_identity":"9713dd1d21764c88c0284f32a64f5d17fe5f01b1ef3e433bada2698ddcf2410d","loss":3.5570104143639076,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"b2711c8b050ccb2fd9819cc6f26a99b8a73497d8f350dd09b1ff8f780ca23464","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"ea76466b6106e1a98bb94a0fae2c8cbc4d76b4a544b891c30198b41a51aeb8e9","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"a6eeabcac1c89882033b6dbd0bfe79639cd7e50333fdeb8f31da88bff6281e14","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"6602ec74bc934352ff43366813485ece546b0b96373a9f16ed87c18b1edad8f6","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"bc2e7edccf64a55d8d74ac3b081571da7b4064aa55220203973d8ac001b641f7","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"ab1c13b85d020f0a737c367dfeb4fa418a7990069d110765457ce2600b071946","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"cb13809d7bda3e52c87df5fd177690e214ed9e3a863e52c93a26114ba603a337","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"27ecda936eed11c2d711aa6660884911e27669a078a85cf2318414ea0b7dbcda","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"265047e39fa3b15af289efb25e5689c4c4b66b8928da929bcb413de40225c46c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"3351097c36002dc321a32d7f25204600d31e53cac3fa49112c48584d4b7af32b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"149c2b9ddb2c796300b9abdc3fb4416e6d26f32d2ee6acf8b6846dde60528448","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"bb9841ab941a52345f3a0bad004084d65d59aaec60201f391d34678c2c3c3deb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"531410faf5912ecbcd3b19322c42318a3752caa59da0f5f96ce00e5a27741442","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"0b2af87d0ab434f1943eccb459daaf405ce75d3ad6c94ff90741ff48d82fb859","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"78eaec4906f0dbf1f31c4be488d7748963435388099b8e3cc0883db1958c61df","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"9c5635b5867a5795d5765b4927982652076d0e7eae1d41a2da24d98f0d9ababc","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"6ef7a2ca20c8eda232435101953148d9a4c6075ac86fa380ae76bcbfc5d2a808","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"3cdf4be6e1fc155b76a9e8e35d9cdaa9efd72e0d8a0d0747b43c80c7cb09e06e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"6a7eadc220d6b2b90b222afd73c8123d1e1bb43adf81b3451d8ff9f560f99493","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"8cba7cd4190c46dba94e1f3e95d4bcc31a97d5fc8a2d327e65a81f77f2e0fa47","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"dd37d022d60e8aa9646518689096c9d64a435d013e2ec31d713b96c66f4d618e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"3396ade1bf20625ce132ad18778b0dbd0aa1d7101b98917c0c13176ab2c22631","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"781f6e067ca1ab072475dc1ea9a89676f202ddaa882bad6bc43fcaffd5a1e629","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"add80dcd1fd882e1d60064136cab7647066e7c40be839872725cf3cfdbed947c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"e97dee90e2c2d05af7d9dc688939c6c794302893fe70d1ebdaa56460dcc53da5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"aa33fb8275845ab2f42e948f38af1a999cb350c3fe1764d512f22cb5f77d9a73","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"d68aef98675992d957212f916083b1d324f34dcb3601fb317137e8df740d0fd2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"7a1a39261fbc65a801a49c94abed61d4dac57ccdd3b976bd691fa4cc9a55b5d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"b9ee46d6371275bab9a3f96966cff92597da9898dc9602767b6ca993e4ae5fd4","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"0ed4914b838b27a529753f89fce262879fea2b43c76acf386bab90cd21aa4912","bytes":5255}},"requestSha256":"7eb62e51cc9c342b6d0e8bc1e6960c4ddce6e5ea0e96c10786fa9af9a71e0a47","metricsSha256":"6f969a33417e9717711a1e4fc76297a52b3574d9cfd55bc2e63e4a21ca4345ce","continuationSha256":"1627dbe0c792ade7db6c2bb5f3838b242288d07a1c5cd2c66754d49858d5b9c4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"dd37d022d60e8aa9646518689096c9d64a435d013e2ec31d713b96c66f4d618e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"3396ade1bf20625ce132ad18778b0dbd0aa1d7101b98917c0c13176ab2c22631","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"781f6e067ca1ab072475dc1ea9a89676f202ddaa882bad6bc43fcaffd5a1e629","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"add80dcd1fd882e1d60064136cab7647066e7c40be839872725cf3cfdbed947c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"e97dee90e2c2d05af7d9dc688939c6c794302893fe70d1ebdaa56460dcc53da5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"aa33fb8275845ab2f42e948f38af1a999cb350c3fe1764d512f22cb5f77d9a73","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"d68aef98675992d957212f916083b1d324f34dcb3601fb317137e8df740d0fd2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"7a1a39261fbc65a801a49c94abed61d4dac57ccdd3b976bd691fa4cc9a55b5d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"b9ee46d6371275bab9a3f96966cff92597da9898dc9602767b6ca993e4ae5fd4","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"41054d82be07f70e2d4bc406766ff8b0cb3126ccfb4771f385273955099a888b","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8f7b2a79acb36ded68e42cadf3663aa961233597d49bcf57a9a000d6d8f675a5","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9713dd1d21764c88c0284f32a64f5d17fe5f01b1ef3e433bada2698ddcf2410d","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2480649","recipe_identity":"cc0b0fb43a4ef294aa2f9a0ffe5cd07bfebccafbba86f0fcbad43f26f35f10f7","loss":3.568927825449281,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"0b5db42e8ed7ddd1807a4d77e4eccac01c203af8ac312ea9990c3ba38b875e3d","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"f64908407bd10480345826d3cee2a7d8a873b73f72709e0a505982f0d97cf9e4","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"dbb05ccfa337a360486ce9365a02be480b895c46315aa22bdbc6e6116202628b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"6ea12ec1b8d748b1511c0d3d9a469aa855c8d7c1c0b253d93473754e16999167","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"0e95cff40fca06366359fb7b4303777d7e9225466f5aec064b79de5d09da9421","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"4feca3c7731c9d313c2ffd5666a105d76ebadf8394ac23d29a2598c7241b8a2b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"5b631bf8d01d1ec44e893ed01d90777988175707117e75932f2f6fe887d46f2b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"4bac3b01930f0e4a63fa117dfd5512526c994598d6f9892ab2515de653bd73f7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"3c93d83674467f3d245502e6ff4eb32dfafb76a0007ca538b128379facc92472","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"87fc3848d6ec111e2ca61e9798bf63e029df64239f8b5f55189c11d305036c4d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"1dd4608de2fb085a7d1389ab4b081283e1723baecc55f5ff38bbf260a3021b37","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"4ab70673699adf4a77c809f8274abecbe32332542098e3fcd78509dc7db9c237","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"3c8ad2d650997db8958e178d49229958e15d8e434735fce5f8c32f83f768acc0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"a293f6c745c89bc00e8585118b3ba7819653d5676d6baf5caf7e2e1be88df8de","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"9f1d2037f6b8f84e1c2b6552dd9ad10f469f8787392dc3363b07e86309d975cd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"69942f7a5a5d58aaf665ea0382223c29f17d421bc4a7f52c8e2568e2ca8742b3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"3f081d3d9f926ca3e554fe88d73919ce571bdd969f359e0667f1e8cc20435bf4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"31df002e2c8d1cdffb3eeed962b386c38909f0cea6201dcbb3646cf314b817bf","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"348dc11fda40cce52f5b7fc8356856e22cfe757aba41ef2546b3ed8c378f1e3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"eb2f224fc64f41941db6e186bef54462b0406e7750a565718cd806bd0052e344","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"b5547b2915b29ea291c712a409d8508cdddb378d7c653bd14529d8834d9d3a1e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"e14979d462a41e8990dea1c49a11db23cf3750877f04b7efd5aaf3cf544631a7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"2da72d22c422dc08d627386fdcf8b89ab55a4a92ae2fce15d0f42bef4e50c1df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"637386bdacbd7bc13a5ad5fcda74bffb22fefb72f20bf3576d92146d9168586d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"61d44c0c1cfe2528eda20e59a115a6ef31d70490367429544463c8645b69e539","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"71e9811caa627eed9ba9002e5ffba9e4363316410bce3473fd71a49c5c5b8cc5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"47b683faf0d69fe8e7187c18dcc47736085abddf9807c2beddf6ffe325ff8e2c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"9ef01d11003a28154f86700edcd56eddf55de76aa6de31fc0be74c1a442121df","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"62fcf8c943197bd57b0d88eb97ccad606e7953a87ce89675cd9797f1c5ba345d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"7245fb90df4c8b2ba1ffe68fded9f8d798d9f65641a508fac0ebd67706c587bf","bytes":5312}},"requestSha256":"115ba17d9fdcb4637e40849d0c3b8aab6633cd30faa74ce44e3ebc8206544bea","metricsSha256":"315b3997a8d29e744e6177265e9e96bc35c3e724ca5c81fe352857fd2dc8d0c8","continuationSha256":"d7a4770d875683576af3c973814f8d26668d23e258d70b01d7665082250ecc66","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"b5547b2915b29ea291c712a409d8508cdddb378d7c653bd14529d8834d9d3a1e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"e14979d462a41e8990dea1c49a11db23cf3750877f04b7efd5aaf3cf544631a7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"2da72d22c422dc08d627386fdcf8b89ab55a4a92ae2fce15d0f42bef4e50c1df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"637386bdacbd7bc13a5ad5fcda74bffb22fefb72f20bf3576d92146d9168586d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"61d44c0c1cfe2528eda20e59a115a6ef31d70490367429544463c8645b69e539","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"71e9811caa627eed9ba9002e5ffba9e4363316410bce3473fd71a49c5c5b8cc5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"47b683faf0d69fe8e7187c18dcc47736085abddf9807c2beddf6ffe325ff8e2c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"9ef01d11003a28154f86700edcd56eddf55de76aa6de31fc0be74c1a442121df","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"62fcf8c943197bd57b0d88eb97ccad606e7953a87ce89675cd9797f1c5ba345d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"576349fdefc629188ab7bdc805c2b3632e51a5ebdb7945705042eb8946961302","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"7de9da81981e85dc44d6f7055902dff02bd98c5427ac496120d6a20373635c05","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"cc0b0fb43a4ef294aa2f9a0ffe5cd07bfebccafbba86f0fcbad43f26f35f10f7","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s42","horizon":40,"attempt":1,"job_id":"2480650","recipe_identity":"8c904158e05ead49242cd34183b9d21a57fbde80b95f3ffdf67ebef3de44ca62","loss":3.4943263633001638,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"35eed9cbc46837d5173f4163846ed35a23cdf8440a3d1e676d8681681dc569b8","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"162ba4991d9f01e9bce11e97159f57077e6aa550bb31d572cb8370af204e6881","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"8e07554c62586c790ac63319ec416cbc79c42931288fd3bb0ba92a4c6f7b7431","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"ac10b172d74a08504414aa1e7ab6e172e88b1296d3f090747e1fc6b5075d602c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"b97898710f13ffe04e8f5f38bb55cbdd7c403d405f7be00e0014992262c3bb99","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"b482260854602359101a5c8240c49f06703c0e99c625b66de1569d73bb90b4c4","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"5f282bf54310336e4059c423b43eef35e996f79e80ab53569b7c5df724f92696","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"9c76737683a00f958529734a5f7739700a902cd1bda602bed2e12a34d278abc8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"46946f9d58a700f4e8edf76f28a18af572d3e518e895829bd38a1b1e5d5bfa08","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"927ba6584f07816adff9a1c39fae601f9fdded777e59dc1ff4f49c9c77012f45","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"bdbe9800e58b149b82e525a550b30ae4c8fac46894809ea5d73c9c238f4058ab","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"fca67c525afb55e374bef1477eeb45f0b5d6d78f71b17a2736b2f34cf87d7071","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"ae014dc46eec7af34e32b8c891890e1e0682a6edd5303231d1a22d8ac8d8e25b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"a46506a3ee4d916c48a80e8d44d428b194cd41d608ca325826db5320b914452d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"ad21b9feb8d3faf85ddf02bf64046a693c62a3cf02f2d6bf44db0c430417e71c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"48242cb1a01f2b8618443d43980b445acb07460e9e4e09719ccf62222748c916","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"0f86550cf10b9b11b013b866115d22d5f828273ef105905acf2a1f859f6a1982","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"4f9ff87890c62ce241f8e7d4fea6ee8f2903245181f22180ed65cc81385dbb87","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"debdca2e1fd0d2f9cf64afdc17e4d0dd79b07a346bc1654e115afbedf19a7a4a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"b75985113ce9bc45e296f2539fd342baa2d8d38bacfb7f303c8a6b3d8f933d09","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"a48c4be34b88afda637ac911b6265e318d29e372c81bf64953c370a7acfbff44","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"9efaa605c71e410e1f0dfa527f11b382c38bb35dffa09bb2ce5d62f19e4d5f4d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"8be8f54addc7d6dd78b19dfac4fc94c66c572aec41fb7b63479766166a8fea68","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"50dd31881e94ada00701080e787048a99648f65786e512d154a27a884c15b1ff","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"4dd7f5ab12da0c4196b6c8a4b2a9513881eaf67315b949b6e4055cda29ee89db","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"ffcbe7614864f23db6f7f17091af0a49f5bbf42ee662ebcbf2d719c625944c66","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"78056e7a79c02d938e6d4904ca51203419e4c02e5580bdf81524b31238f98ff1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"85f17bebdc323fbd938f42b39e61f18fc59e1a41dc7146b88e25e5557a9c0537","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"96eb8e30a5da1f0557903c94646ad59879bd7bb47ffd421ddf914c83996461f2","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"1267fe6cde9631a144fee16dee5df7798fa8b44cc8de7ad437a9ed0601c2053e","bytes":5341}},"requestSha256":"fe9f138d065241a8e7f0d1642a63f7f5009e386da713f8a1dd984fe5601b1781","metricsSha256":"35ab5081072b9f4d7603838d6f6246b14c9118535d4d68216576d65a7700888d","continuationSha256":"d0c1b8c2a6d1da35ac237fb7fb48bb3f3e5dd797429625aad903ad1ef06c70ec","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"a48c4be34b88afda637ac911b6265e318d29e372c81bf64953c370a7acfbff44","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"9efaa605c71e410e1f0dfa527f11b382c38bb35dffa09bb2ce5d62f19e4d5f4d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"8be8f54addc7d6dd78b19dfac4fc94c66c572aec41fb7b63479766166a8fea68","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"50dd31881e94ada00701080e787048a99648f65786e512d154a27a884c15b1ff","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"4dd7f5ab12da0c4196b6c8a4b2a9513881eaf67315b949b6e4055cda29ee89db","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"ffcbe7614864f23db6f7f17091af0a49f5bbf42ee662ebcbf2d719c625944c66","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"78056e7a79c02d938e6d4904ca51203419e4c02e5580bdf81524b31238f98ff1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"85f17bebdc323fbd938f42b39e61f18fc59e1a41dc7146b88e25e5557a9c0537","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"96eb8e30a5da1f0557903c94646ad59879bd7bb47ffd421ddf914c83996461f2","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","sha256":"4d062d497851f545f8a64c57964bf9c65f9560ac7b82ddc0e3bf9f5833d6a112","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"06ad20af0ff3ffbb2fd364f492421711810cc1a4f80afaadb39c7e2cf3889aee","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8c904158e05ead49242cd34183b9d21a57fbde80b95f3ffdf67ebef3de44ca62","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s43","horizon":40,"attempt":1,"job_id":"2480651","recipe_identity":"b1d844e65cf0c89fb7933824a07ec7becaedc61aa4025ab5949b13be3cc491e9","loss":3.496827753624678,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"3fbb1153e3bddbed3a79169731dde2a593147e27f62f088da71836018995366f","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"d6b37d32414ec591a33ad41b301d446869bed442df42d0067983d57737257a35","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"3da601745c5106e0bece33a6b670691035b6871280989728eec96cca1aca448a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"a760c119feb010316f373493de5dea0fe19107c0b7d064665c74ba06e54a8264","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"000427b7b77d4112597ed884c5af66bca293bd1372d52c28e21b77d440a5a1e4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"8185556ce0ad29173b4e8df0bcbe55a0f73c6b61cf4a79676db8f3f61d4fb11f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"191f212ec1becb53bfff97c9db1b95ababcfe7b138d754c089c3cc4f539c0208","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"4d046b720c84713b8764c8720459ceea31e0973a7cabcf04df91a42bd692c84a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"a7e99f45b90035d75f3ffc14f33e56ef8fb3d60df6844165443c7237c96e2140","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"5ef2c5a1f6914916ec981e8625b023b2650de02d981b4f58a24ea50a48a9cff7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"5b3c860098cb040df0e24c215d675f179bb455883c762c8b00660751a4ac230a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"d59b094d15069e43c76f81fbb3f7799c3d3aed3e0774bc76468c547694fcffe0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"45b10f36c34ba6f42baa221a7c02c93d2d12302a2e874a4efde2b874fa2ba23c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"8df8d9a5ddb292d59977f7ea1e6d02089fbb15e1ae05cfeaa7638a3fe2040737","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"755226eb7ce56f3884bb44d388b357fa6853bb92b4df40a759bc7319ddf14c55","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"e5e233d89c90373db53e4b975f5649c3bd699a236d378e308352d3c39bf32c38","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"e5dae0202d28b2171c5643822686840e520225700173f337a8a1b45ed4d5f328","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"d58e54b8928d4add2b0a207c7e4859095a3a68150730a944d387d54d2eee946c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"bae5330d92ee2d17ca391ffb9a1f33636b47b64bfedc063f46e6ed01c759c520","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"25a3ceea413a0b0254ce45f5df5a77cec1710284aaa3603890824a18b7c483cc","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"f0baef0de4cda403b45454cbe33d23577bacc382a4f499c0c7a9e3c0cf5eef8e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"34fc00f4f2825c721b661067fb2b9164fdb3032535ef5eb32f567052a6a4f445","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"7b7f22ce340e26f299f71eb707acf6d79d987827392e7821dd387e3a457c8c7c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"cd72b7b7f4fcb93d5acf93d0550ba03ce39563c06e8c912e9a6cb5eb84acd258","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"1651107d6328e89a5fec8f9e3ee7377684d278a9e6b204d96236a1f75068e774","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"521e7b042a7fea94c90d6f4e38ead101d00a0a4e334535eaa77f2cdc2c862a5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"a6a6ce58190ea51e7d82ced6de2cc9f0b580544517bb604cd29b66e0ccc1498b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"be62b8699c11b27a1ecc2e410cadc097de700d556a74d6cd197b18d4ede38fcb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"c3c7d7e5f010e28f61240e311d0198229e4fff7f3790e7552b3b0032d646d052","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"9fefd3adb2b05d0ae743046af79b5d7ddfa03a98b6ed1c89ba1f65ee699d3aaf","bytes":5253}},"requestSha256":"438964324aea140a4bb2c00c08abc201359c1475cd6a79945114dda2d60cc19d","metricsSha256":"8f45bc45deec9572fe9cf17a2a42a853687c10667940d3340716780b1d7d637b","continuationSha256":"e0fe1acad909c00d972f19b4f55a3a710764ca5a486cc8239f030e06d05029fa","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"f0baef0de4cda403b45454cbe33d23577bacc382a4f499c0c7a9e3c0cf5eef8e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"34fc00f4f2825c721b661067fb2b9164fdb3032535ef5eb32f567052a6a4f445","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"7b7f22ce340e26f299f71eb707acf6d79d987827392e7821dd387e3a457c8c7c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"cd72b7b7f4fcb93d5acf93d0550ba03ce39563c06e8c912e9a6cb5eb84acd258","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"1651107d6328e89a5fec8f9e3ee7377684d278a9e6b204d96236a1f75068e774","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"521e7b042a7fea94c90d6f4e38ead101d00a0a4e334535eaa77f2cdc2c862a5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"a6a6ce58190ea51e7d82ced6de2cc9f0b580544517bb604cd29b66e0ccc1498b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"be62b8699c11b27a1ecc2e410cadc097de700d556a74d6cd197b18d4ede38fcb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"c3c7d7e5f010e28f61240e311d0198229e4fff7f3790e7552b3b0032d646d052","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","sha256":"ab05d0df5681f94e34db2d6285caa50d601c2560ac9ac20b8356062326cbe6fa","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5723a0bfb8869dc896b9c9e3e8d79ab87ae77f24d4df9dec52f5f4c31dd2ae50","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b1d844e65cf0c89fb7933824a07ec7becaedc61aa4025ab5949b13be3cc491e9","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s44","horizon":40,"attempt":1,"job_id":"2480652","recipe_identity":"364db1f3c6359689033ecddc4b42ed28136ca00140d40ae269c0221606520b1f","loss":3.495527070617188,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"a5fe9cba10dd7b0e99d4008c25d033e4fbf3dd9f8e3cea8bf5ea64d5b4159213","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"f0ea584fba3f312f8384ed4d4723fdd8ea5f24d52f7be911a26052f94c662d9a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"9232e8aa052c577998297d120c7ab9ff7c41f46abc633934d2fe619bd690d79b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"1101293dde86811d7afd6f1d5520749f2c683fb981aa1c97e1b4f8d3cdc6b7e5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"0e4796cbe04c1d518d8df2d907cf2dc05cfa6de80557791b949c2d3da92e65a5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"737ba273ed4c88f924e600cf7b7dc675f4fa33be0cf994a0327d3ad533de5301","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"e3b34ba8115033375e4148a8b34ba42ce5f7ebd5910bb7e367d4527140de60aa","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"daf2402cedbe43c7650cc957686dbf6d977c076b176f3166f4d76822cc32e32e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"771062d001d240587665a1ffbb30484bfb1b038ccc110d4cc258c65b6445df98","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"a487b4dc98f4253f74724a905c1ee506ed4178476cfb3ca5aac74e9b8ec9cdba","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"24746108487ac8c563feb659cc9fa4e339f8639bbcfb958ca0fbab4ec0a15733","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"15bb3dd088a3e8c2e657b3e6e2243392f3e968940956d20365c54112a5c0bbec","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"7fe92f72591774da2c3d6b18a662766a8a689f101c9317bb2e03259b129f5729","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"e8fc92d491b29208be903fd11c21fb91829ff3767d0c341a4f64a69f4cfced19","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"5db29c39ec667928e5a82746eefff8228040e793b3362036c71e2899623948be","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"48d3f1e886205a9fa5ee6c8862ecd45ec5592ad087d2a61d1267cbbb820a0d01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"ef68c0ecefc817ccd47d2075d2017609e25b4c4834332df5c6691a1f6d628c24","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"939686fcf144175e9c5233026f1fb9b087ff9ebd5cdfa86a9e7f4e5dd63612ba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"5a1e9e03ffdb6efbe51b074542a62b3b743c4161b4c0322307e9de493e217d9b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"10f27352c43260d5450c732a55e9f58ece28dfda81a34e36667a9e8d8931d5a7","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"f70e624f5d4ac6dfc889004af2761d13a0d1a2c355dd595e065a0ca6e30e35d7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"394736e8c05552b6c187b7a41340e54d9d87745a5f3603108ef62a03ed4590b6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"c583fb94e8dba7b121d0be0ebd638c11c3ae00e5c4d4b0597971ef6e32b59a14","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"3014f00ffc83d7043d7ba468b999cee7eedb00c81ec80f5d49e376d8c6ae34ec","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"41017981457cb5316dc2eb5687e1c5dbd3f375c15b3b3208eaa0ffa3c3873ceb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"f09361233f1347836e78923f7b1777b87dac9fb4f6bb9caa330dfcfd932800d5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"94859e1acb165260b66630b156a2dbc9fd6c0fcff10b6e337d41d1550c96e415","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"1b05e519d2029806f14e9d924d0730b8abc2e833d666b7af137808c46cf78ee7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"cbf7b02634f2e5ba60c2e5172da033f265fcada45af3608e9086c6cfbd596034","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"30a9bfd80e27ca6fc5570a4ca23fb24077cd0d95907ffe5336043ddae259f736","bytes":5310}},"requestSha256":"ddc84a2debd915a66c4a8dd1e8b57b3ab1f5afb66e554804528b519bfd9b9caf","metricsSha256":"df3a628da2f8a947f6fe1c029b25a03d68db158774d3f7ce699a35051702ae46","continuationSha256":"18edc05a7d0138bc7c5a651caa2a3e3eb5e430f62314c63d8ad2022e7fdba85a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"f70e624f5d4ac6dfc889004af2761d13a0d1a2c355dd595e065a0ca6e30e35d7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"394736e8c05552b6c187b7a41340e54d9d87745a5f3603108ef62a03ed4590b6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"c583fb94e8dba7b121d0be0ebd638c11c3ae00e5c4d4b0597971ef6e32b59a14","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"3014f00ffc83d7043d7ba468b999cee7eedb00c81ec80f5d49e376d8c6ae34ec","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"41017981457cb5316dc2eb5687e1c5dbd3f375c15b3b3208eaa0ffa3c3873ceb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"f09361233f1347836e78923f7b1777b87dac9fb4f6bb9caa330dfcfd932800d5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"94859e1acb165260b66630b156a2dbc9fd6c0fcff10b6e337d41d1550c96e415","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"1b05e519d2029806f14e9d924d0730b8abc2e833d666b7af137808c46cf78ee7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"cbf7b02634f2e5ba60c2e5172da033f265fcada45af3608e9086c6cfbd596034","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","sha256":"d35787de665d05e78d33c7d746daafd7c5048a605a5688e6f4f520c7cd40824e","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"68caec269fa82ef5c07ebcc94335bfa05e89c7cd8a5e658a7dc35dd0f4ec9df1","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"364db1f3c6359689033ecddc4b42ed28136ca00140d40ae269c0221606520b1f","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2480653","recipe_identity":"ad987437e6b33209382bdba23710e07151e0b22463653cafc755d92b2c162b16","loss":3.4996621550603053,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"8b99983570644455f654ca0145e84318941101a1fa5b34feed2d94e00aa2cfb9","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"75c133b674548e496e1139bc4a7e749d2e4e61688fe2ea8afc9063879769dc08","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"27a2089b19565a8d376df9f157ef3e4420c7563a710526b5844a2ba4cea8ddea","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"07cdd911d121f7b3dd14cfca38f7808236b1c2534f79ca33fd22da23256eb336","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"8d7220b16e8c8d44289f33346d75a1358df17fb11c3c21800a7eb5ea3091e101","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"8061001d7d5b07e3ba02bf654b072bfe4f359925cdc94cd0e26475ef099df122","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"84a3816d53a9592baaa022870ec44701d4a55fc69aec37be3ead2fb142cb3dc7","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"1ac0ac5bd1874be10ba7afd296aacc499683c59db72d7fc7b8df275c0306ba63","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"fae3bb3236cfb7c84a9d8c053fa48efaa030cf7b97ff9b130788068847fd9d4c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"a870eb1f0d1b66d9ee71a35658e9b0258549545f95fe7d5c86020a5828e12868","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"d7b025aa1806eb93b1b9fca69f4f489b2d6c24321cdf7a987f42d899178b8a1a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"862d2ff8434c29109afe76addd28a31b5024e35e5c812f2ba0a67fd76f39e2f8","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"a1af1116c45fa2267bdd0b0d83ea95815a363d5b8b864d47c8040fd51ce7f81a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"daecf67f9e55487aa8a8f41d113d3a9d820d0c6e704eb6fb9f580a5b9658b61b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"e1cfa8e60e799e143091ea0749f1c86b7db40c90be4f22c8c916c6f8a99aa664","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"8ee746a3c92955c432cee7abf2d345851350d59acc91f04088c98775b8b95f56","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"8e1dbb686e516cf24ed1287f412f1576a34093f247fffc92af3bf87f93c67e5d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"b7619f703cd50461d3c7cffba856ec1c79826409d08847421ac73e54c0048569","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"f50bd495a5d6bbed8a345b912d06f7a91be7e13d3e49f00f8e3d6bdb017a37a8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"a4d3439cf2c3c04ac0c8ebaf41b531798472707bd70a39d666385af0f72725c7","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"da1db943fa1d626cfcd3c819037243248740f9f76be7dca0129d75537a35d4c7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"801ed5e36168e7d871dc429bc3cbb0d434e73eb1197e884f9ab4d50aa5fa5ff1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"1ecf69b793cf1014f14bc2d90398b121c7ca383425256a2cfad346d3ad58514c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"75822046ac41a763a101486177d9e430e570c97265d3edb81138eb43e6735a27","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"0e6828a08ecc165f86d3c5a82b1a31179ad5426954e4a5d92a6e31052e0166a8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"3035ff364ba75515d773f17f151f05058ef07113b55a6c8a73b0d5e10ed5c9c9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"99eb69a532e3491aa50d8c8abeaacf1eacd5423b9b5adee3d7df2bca1243351e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"5aaee6ff692df9b5e4cfad6f5e765b82cab4ac68488011d3f2dd0acb88cdf299","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"6ef178626deb317e7a9f0a14d6011d7cb63eb0f1e4f668cbc1843f5e4cb7e1f6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"8c0481bf2533be3f3bb8c2ad49b6ddf8aea0b233055a8b50e8a7b5cef65b90b9","bytes":5339}},"requestSha256":"b0b9a26a212277a5605235c8200f5128240295eb8f6481f75329b6c8f64e6363","metricsSha256":"c70120cd9d25bbea716dc2cb3753437842b34a8005e5a8cf6be8a675d1e8d70d","continuationSha256":"f0a0c5eb55fcef847a5cdfb3b0862794a114e07ea25f01d818de8ff9d8cd9009","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"da1db943fa1d626cfcd3c819037243248740f9f76be7dca0129d75537a35d4c7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"801ed5e36168e7d871dc429bc3cbb0d434e73eb1197e884f9ab4d50aa5fa5ff1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"1ecf69b793cf1014f14bc2d90398b121c7ca383425256a2cfad346d3ad58514c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"75822046ac41a763a101486177d9e430e570c97265d3edb81138eb43e6735a27","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"0e6828a08ecc165f86d3c5a82b1a31179ad5426954e4a5d92a6e31052e0166a8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"3035ff364ba75515d773f17f151f05058ef07113b55a6c8a73b0d5e10ed5c9c9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"99eb69a532e3491aa50d8c8abeaacf1eacd5423b9b5adee3d7df2bca1243351e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"5aaee6ff692df9b5e4cfad6f5e765b82cab4ac68488011d3f2dd0acb88cdf299","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"6ef178626deb317e7a9f0a14d6011d7cb63eb0f1e4f668cbc1843f5e4cb7e1f6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"3bad7e4d08db62507acfd3f667b6be90e893ed449ea7007ed4cd184eac5ca986","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"781180035e2e2711e98ce9d06f866738d71de2f60fe24d580cf05e3e1d7e42af","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ad987437e6b33209382bdba23710e07151e0b22463653cafc755d92b2c162b16","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2480654","recipe_identity":"9b653e8a1d33580b8b0293b5e799565b12348e69e1d31ef18ac5f0fd23d2f424","loss":3.497683933870879,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"6257e77fb70d780011fd7c8f11a33a843ab2484ca3c067f01bedfe6b230464d9","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"1b913bc4e99f661f8e4d39c83752b797f6f8ffc5af73d9953e397fc0f818526a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"482d4d62f1da243c946187d7681c4ca5658479c6281881ea849b99c09881c431","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"090e8c9cf93e9f2a5cb249a8f70094471df9fd7179b4e3549f3eba3e1ff3f7bd","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"8599782d5ce91b46b7a995c9d9a4c39c525e0216fc1efa13c6e0a9f1bb7059dd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"7a1c887516b299ce02ebfc77a5894773c5fcaed71bb3af5a8302da43e31e2af5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"e28420d57ee6de77fb5e7742e6fb68d80ef4a594986a8ffca90f59bbf0f297c4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"7e014ed89b9ce14b85cb8e276296f406501aae2b61bda0ffdd9d2cf15b35f994","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"5af6051d7235d6a46a3eca7acadc07bb17986b9762e313171b34d4cef94e585d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"ef21847413e4f8bc88fac251054eb8481da68c7c15b626a8cc177dd009cdd905","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"3338cc371315b07b3610de80e2572f7c43169753602b8ceb877a64e411357391","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"e2c684069a82f94d86025b1f59bf8a895e094827cdd076e83b63c8c55c62f07b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"d19bedf8016b3d3b94fa92f05e4c1a157ed05ebcdf3e8b6a0861ed941039c849","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"c98d1d6b40f641d6bd285dfeb482ea99f528aafaa8f8c8cf0e85a66a7addb406","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"262976cfa18a29388a8a4f14dc5b705cda6c26524acb32176f1f2dc97b6387cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"79f27011e12286380596c77d8a2339708c887d8b9d2cb43b2d7efc7958cb288a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"a4df207e466ceff6fb97aeb860b74b8c9896eb40989b796b7368763f08ea7c75","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"e4e10271098cc1a8f1419d16b0b7832e6df36ae0875bf99c24421abc70ae5980","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"738b5732bf7cf497c89507d2b8cb0022091d1a629050f1d7eb1704bfdc6356bd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"cee1ef8f020f32db81fba881ab92e836581abf29762206b2aaaaa095f51b55a0","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"aab6be5a39ddff107d8b0d11220a2274d10f80556081afec3d86e0188bb5ba06","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"f32313af8f8a8564ab9457a85848cacc694418aa05524ffe10031a70e5cfc981","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"6f107bddc7cf5df6d5dd04238aa89b1228576bcef76d67d7924bb1a0e203e7ff","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"0ce43fff10632e4591f4bed483d9e25ccce62fcaf388ff9c7e95a3bdc0c54115","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"2c3df5ed1fbf3ae6204cdbd926bedce47db4688a89781180dbdcd3a3aff7077d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"fa4557757d82c726f0c8418cb3df5233deeb6696d9550eda6c8c74864a4cfa9f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"02e2ed6b2feb73dd4d969a98739ecaa0fc78976f1c4bd0aec8da1e3bfe522e8b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"5468d3d71664e3c29c4421668c4a6104d56c70494766c08b7f4b34d2846dacd8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"16ddba3a0184409307007f36110b1b01c25cd9e1b42172283693f5537552277b","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"9427d5816648d28dc5b638ae990cff15b8dcc9bf1d6b642bd05bec52500430e1","bytes":5253}},"requestSha256":"0fb4f0953d1777587f53220e2f5cc3c5150cd9d6f5ca18fba5554a98be13e745","metricsSha256":"c2c7c3ff5887bbf0de1ecc8c2d490c117ede18d47b4c47d8c77f7baff1ca260e","continuationSha256":"1bc1a85e52592d1838ce638f97468e2fc84ca718d26578611a8004a27e4f570d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"aab6be5a39ddff107d8b0d11220a2274d10f80556081afec3d86e0188bb5ba06","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"f32313af8f8a8564ab9457a85848cacc694418aa05524ffe10031a70e5cfc981","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"6f107bddc7cf5df6d5dd04238aa89b1228576bcef76d67d7924bb1a0e203e7ff","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"0ce43fff10632e4591f4bed483d9e25ccce62fcaf388ff9c7e95a3bdc0c54115","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"2c3df5ed1fbf3ae6204cdbd926bedce47db4688a89781180dbdcd3a3aff7077d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"fa4557757d82c726f0c8418cb3df5233deeb6696d9550eda6c8c74864a4cfa9f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"02e2ed6b2feb73dd4d969a98739ecaa0fc78976f1c4bd0aec8da1e3bfe522e8b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"5468d3d71664e3c29c4421668c4a6104d56c70494766c08b7f4b34d2846dacd8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"16ddba3a0184409307007f36110b1b01c25cd9e1b42172283693f5537552277b","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"7964b93f41d4fa405125fe90a25738e55b706db7f6d4b28ccec3d559d3c05685","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e479db68ff6798c28fa3580719e409bf05295d69a39f209bdc4662970ec8ad3e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9b653e8a1d33580b8b0293b5e799565b12348e69e1d31ef18ac5f0fd23d2f424","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2480655","recipe_identity":"a9945daf6cebba4f76234027768a16c6d468523c119b4a67217f540300a81739","loss":3.497369623464941,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"00f551c01dd1f5f5c1d439ae54de2c735ad5dbc8de66cfe2d21ca0434dfc474e","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"b822600501f21155c9a880bde21a2728a565493325281704d191a68680395010","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"d83eb8d0fa899b78006e2eca352607d36e3c245ab47f8aca9a63028e5bbdd3ab","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"66a5f192d98748106426c52bc12c25f6098c96dd02e257055e914599f8712bcf","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"0ce53f4ad636084aca47984797e045e79b61ff2c5e22bc4e97c0ba9c7b8d94ac","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"d3f86cf92bccb720be349030ff9254bccf39233bbfbc47b27d5e419c000f7c80","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"b3a03a73411501f71a34f5588f7b68e84c285a1f6694e4e8d199569e7a1505f0","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"8dfe88d20c27a69b8b564351492f23cb5c66cab3ed5e020759f7990b3420ea5a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"4f95fc74bc2eb6f272fd4d5e3f9a359cbd948ed1cca3fe0f65e348748d1390f6","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"492058a1a1f387d9045a51301cba1ea997d0a3f01655d3ea58f9895b5806f8f7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"7642bd336f0fe0d8c3fbc5352f3d71b70fc0c106f58a272cf6d838d3ffc50284","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"168684e24c299867c742b0a70aa06974e0c7df59e8861497c1c4b32acb796a6d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"b3d47151983ccee454e5ef33497d09034b7c99c47f4e86f3770b8611ac1ce5d3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"6cf290c3e1410d1375727635af2847667277cb87f3fc4ecf2b6fb58ee4bd2448","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"1e259f888884bb34f22da95c9fa7d697b015d5aa286de6597e2104ff6fd4147b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"152e83359e9c71b2ad340a759c60907bf3be68f6c4e189ab6d282902aaf6c7bf","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"4e82c8ca2da3b837edfff423b10165002bb38ab4a0aa697ca8e9c04488b1870f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"e7f2dbf169c9bbd6c3dac343cea72b114ad81e4ee312dafcdef79bb790262e41","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"624ff865d00162b4c58f9bad7420dc41346976eee3d78c3d207856aa06d6ad18","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"1009cbc67e0a3fe25c5ccc86dc2fd4b8db6207ebefa23ea278ca2a6f321e0c90","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"81228b4b78309375e3956815d0b4aa87a747a1cccf94475322132b3b2c2f1f8b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"e8bb668f0c36823029bf4a1f04218032cecace0e9047e0170f9ea4291c26626a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"dd13de6bf8d08a263de8834c7391f5ecc19943a7f5642859deb5e38459e2d9d9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"13761934223b5f9d675a6738857254f348617793554712eda395e25701540b8d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"4e47a983a55f424c6c389168ac212b16921f9b3d70c497bdcd44808c5c2999f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"f1d1adfc88d70cf92ef07f768785f169b7b927996f65930b56d8dff35ee7fdd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"de03cc031ad2b8a230fc439acb9e2bb5076c7d43997f14d3b4c1f654ee5c4153","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"9d8e94d8544933b7f925706b79bd92adb7b84995bd595f196404f6a0848266c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"460dac899953bc648e56684d4cf05e0ead2fe0a8536a33cb8c007cf7a9e709bc","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"faa8b4f2d5d49c50393055e8bd3c8ff11a6d04bf51b53db266f31c59567c8d54","bytes":5339}},"requestSha256":"47dd4c20cce47f6fd8026b5078661ff4bde92077d61b1fb36ea40dba1c24747b","metricsSha256":"378b8b55d8f2e887696235b1b1bf121f19b808ae0ee6e050351269e8235280a0","continuationSha256":"72cef1286d29cf692c166c95fcfc2e09667209d6296e029092f3cb9805f90f32","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"81228b4b78309375e3956815d0b4aa87a747a1cccf94475322132b3b2c2f1f8b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"e8bb668f0c36823029bf4a1f04218032cecace0e9047e0170f9ea4291c26626a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"dd13de6bf8d08a263de8834c7391f5ecc19943a7f5642859deb5e38459e2d9d9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"13761934223b5f9d675a6738857254f348617793554712eda395e25701540b8d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"4e47a983a55f424c6c389168ac212b16921f9b3d70c497bdcd44808c5c2999f7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"f1d1adfc88d70cf92ef07f768785f169b7b927996f65930b56d8dff35ee7fdd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"de03cc031ad2b8a230fc439acb9e2bb5076c7d43997f14d3b4c1f654ee5c4153","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"9d8e94d8544933b7f925706b79bd92adb7b84995bd595f196404f6a0848266c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"460dac899953bc648e56684d4cf05e0ead2fe0a8536a33cb8c007cf7a9e709bc","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"25adfe13998e7f01f949554adc1837b42b6d4246028b8a4a68a4b64dbfa000c0","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2e0db0f9e70fd9825b791ee2c896c70483b784534d7244e5bde733968d3f8aa0","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a9945daf6cebba4f76234027768a16c6d468523c119b4a67217f540300a81739","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2480656","recipe_identity":"78117f741c30ec87e30424f4532f1cedffdac6668d753b64e01deb13106cf933","loss":3.503650241445448,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"f985eae22eea814de89df0e28e25b7aa9455633878f26596a429fae76dbaa027","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"14220829473cebb330809b925dbc3a61cc2e0d42e7598640fa92ecc87c69f6dc","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"d98a4fb067b069d0a685fa03b0d320932bf97082241a4b03c045092030dfff2b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"9860fa5527d29590a6d582a0f713381e77627a5c2f4f23c4a52bbc27e6fe49be","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"e5dff3ab45fffdf2137c1a6c1ea920fdff5cf667426913609ca69605ef6695b1","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"1800dddb5114af209dadf985d233403ca350434064bc3ea7d21d34c2034ae8b1","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"6055de2724938c7e301d4d38a0b206d3aac5b1ecb037ae20a498b436658ded52","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"610698fe0df2901155501fd5cabb8b592933a9dcba7f0b2fefb6ac51dce881ef","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"923bd2a5b1403b1b565090225a28d2f584bf7deca59c1ddb1313bd888b3ed7e2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"e9d413121b7d7ffa9427eee83a21855544e806e4942ce1ccb96d7c0cecd6f0a0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"bdbd09da5848dc27da0aed6f907d1f184007a2007696d5cfc61773b6996c7ea7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"98cc427a85c7585b3b896b49edd8bc6d7f19cd88cacc5d5fbe2ff81e86fa8b20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"fb9c48813563e750ed7ee6d9fa76fec927594db318986df8e806b5ae37c798bd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"19519ddd02a1303e99eb7cfb0b7199f3d6f7a5e14c7e07a7023d220c52ebcef8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"b8239339a07f6682f4e7ee7e928f0dc0e4559020ed30011709f2cab70baec3f9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"88d97ca5a99e2fbade5cadd88137830a6c406ff9fef704a499c71d197da01f95","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a7425b46fe17846ab76a9b9aadeb70636fee0a95e94940fefa958010a9ea3291","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"9339a7e91f37f3cb53bb62ec39eb831e6f50e59538efd586c4f5f0faac0a151d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"c7e9967dbc87dea0035d09e0d458c337ddcaed29ad7062d477fc427daf9b37bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"a31f1952a803d1e239fb831d22ac2cff705bef5b1c3b7230befad5e5c4a38f81","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"dba64f0a1eca9f334dd02a53f6ce76278c6f19de1fab6d115ad227f1cef62622","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"d705e0c50621c65fb3a018118eca92602e5f9dd0b70fd18f86a4625ad66ccc0a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"1213969a3dad5befd45968fc27c9ef81cffef0d08df00b52bfa3fde3c98cb37a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"4150d17198e69346f7abaa9c2b5777739432b3a44ecfbb73cfbfc6d2416923fa","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"56ce7fc6a93c33945c77abb69f9edd0b4431f16e13466d2cfbbe12f27477e22e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"50f426d0cdedbcac270f59d003ef7c3c71ecee74d39f40217df20bf5089cec57","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"adba6a0d78c5935ccd5b1f39b6c4ffbb9fd6bd29657da23bfc25725c0fa627f8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"b47e97e61c56bc51593e436d999548b2e1fc9b1fdf971ccb96389cdf362d8790","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"15fea774d551a32626219c873faf7239bb6b4552610342d41626de6938f9d643","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"3787e339fc559ef0208a13c42b28aacdff6dd505c608d5474eaa049c27a4d6e2","bytes":5255}},"requestSha256":"00511171c1b57a0d62f699684da8d3683e6b56a5d9aa6055ea1e0f3ef83fca8c","metricsSha256":"a95305e536f2b07076f796d966aa7623376a2ab984f77403c88b1cbe3e4aa44f","continuationSha256":"95e7aaa5c003a256cf2e51a953314950ef0ef2e9590243e03267d60a60c68e0a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"dba64f0a1eca9f334dd02a53f6ce76278c6f19de1fab6d115ad227f1cef62622","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"d705e0c50621c65fb3a018118eca92602e5f9dd0b70fd18f86a4625ad66ccc0a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"1213969a3dad5befd45968fc27c9ef81cffef0d08df00b52bfa3fde3c98cb37a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"4150d17198e69346f7abaa9c2b5777739432b3a44ecfbb73cfbfc6d2416923fa","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"56ce7fc6a93c33945c77abb69f9edd0b4431f16e13466d2cfbbe12f27477e22e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"50f426d0cdedbcac270f59d003ef7c3c71ecee74d39f40217df20bf5089cec57","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"adba6a0d78c5935ccd5b1f39b6c4ffbb9fd6bd29657da23bfc25725c0fa627f8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"b47e97e61c56bc51593e436d999548b2e1fc9b1fdf971ccb96389cdf362d8790","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"15fea774d551a32626219c873faf7239bb6b4552610342d41626de6938f9d643","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"1c1b3da84c1ab01b9e451f974a71741da5debf83c1a002188e4427977ff0121e","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"28f0a2c5daa3a77c6456d0dc46c63b18f45ec3682f820bf7cb18235c839089bf","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"78117f741c30ec87e30424f4532f1cedffdac6668d753b64e01deb13106cf933","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2480657","recipe_identity":"a0d9ef891426920a1f3ed909503ea2591418cb5e784e86f9c3da53cb0b331fe4","loss":3.515859640584827,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"92f9e380f0838ae4159a5fa29a6c7172a6ab3fdfcafe51565981ea37d2ea5418","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"8fef30bfa54c4faaeec5714891ce6d0e25867f4514cc66eedbe922eaccc1a168","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"918213b6c69e9640579aa05d5f589cd732facbc8cf8dc2725b3e1f1bb6d9eac9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"372ce493daad01f8269c189e17ddcd309be1e64d3b248bf6e002ff7506a1b3ae","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"76cfbc1b961eb1d1322da104ee60497f79c85daff9ee84aca4c1f7d6a0e549ec","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"21fa1ea7c892d2e355cb5b8da713156b85ac666c253797b6490f0710f5315a2d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"19e83177f9d9d0191315dc483703317774f57ddde3962cdebf55c4d9a34f8cfd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"31f0b7441cadcac9e88019edb8c7c499dd3de410d5a93ed4be01a76c87c7d10e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"2428717b6936c5fe9c81e3f6923a2688b4e798082dfa612f689d1f1f57e72afa","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"512d632dfa989c86c4835f86a402dcff37886374d936cd24afb820b986507211","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"c072ac3a343973dd8036053cc9ebf11404c4589c9e8f58259e7bb314b31a22e9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"83a0a311b4ceac8c1d98f98305f6026fe41768351109ca1e0872dfc904e2c120","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"0fc86f701a5613dda0c8d0ba3b1fd227d65593ff9ce7785ed4c58c19c47661e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"3211efde08f882f43455cff389b94b6aaf14f1d91360e3342890a01b22478d5a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"9178b8253b81050c00a7d431427b7b6b541e4582dde5ee4339a68fa2d4b0fb73","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"b60ca255d9ad97b3387da83c23d0ee629762979bb9d8be28f82a5d512d871db0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"59c924883f5321e7392682bb4976bc3660a40054c67317d50bbc7c5edff7942d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"7049eff3a946d71595c71a6df67bb48cff551427b65156e902e43c653699ba60","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"c2910d86accf89eee132261b3c70ba6343587436311b2e4812835e1211b30fc9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"a99288bd7bbcb676046f155e04c21e6569bccf698fe01119763c5946744b83a0","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"210138818057c5cffdc777f4495e9fe42efed322dd618788f445aecd4b261c05","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"9a62a074fcc78e3c50b32b4c720b04e61be51b9958c4232bf9d740473fa647ad","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"cd650590d76010c3d5eab19248915e29645630cf582591eec44f3bb89cf48f8b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"a7776af3b9556a32bab441eb195ad2011a798bb802018a24b4503538ed10e147","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"7d9568bd31bb4431030c6da96cbd39fc1befdb02334fd7e3584af2a8634f633c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"080d9a57ff0208892d20219eea7e1936042a9931e817da8ca6e007748dd654b5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"54751448ab6969040bfe08b7a096e0704746b8c6a8a6a247f0d2e6ef44ebe36b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"0b9745bbdcece4f56b7b9b07f8f4e3361689ff22cd784cc2d810b3754f69a7f4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"49cc6c5f7cd3ccd0a4728be406056008951943717fcf69666cb6dc3baf9c83a1","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"f2dd9aaca35d0aa7cfa7f906d215055855bcfb44f8ea6bb22bcd3d442d1465b7","bytes":5341}},"requestSha256":"c17eb1913be2543abc033e21b9c8fe95cd8c6c95b685f8f61dd153eb6c7936d7","metricsSha256":"e2ca445db3020fad1da04b2c224362e1b1579fa08764edfe87c4e42c02e4b57b","continuationSha256":"e158996ebcad7a8ec5927e91e2c0d3aa86d65c9044011f5e8dc0207c30b1fc7b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"210138818057c5cffdc777f4495e9fe42efed322dd618788f445aecd4b261c05","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"9a62a074fcc78e3c50b32b4c720b04e61be51b9958c4232bf9d740473fa647ad","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"cd650590d76010c3d5eab19248915e29645630cf582591eec44f3bb89cf48f8b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"a7776af3b9556a32bab441eb195ad2011a798bb802018a24b4503538ed10e147","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"7d9568bd31bb4431030c6da96cbd39fc1befdb02334fd7e3584af2a8634f633c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"080d9a57ff0208892d20219eea7e1936042a9931e817da8ca6e007748dd654b5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"54751448ab6969040bfe08b7a096e0704746b8c6a8a6a247f0d2e6ef44ebe36b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"0b9745bbdcece4f56b7b9b07f8f4e3361689ff22cd784cc2d810b3754f69a7f4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"49cc6c5f7cd3ccd0a4728be406056008951943717fcf69666cb6dc3baf9c83a1","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"ed15d6b3490b783458bc065c31c03854957cf3dcfc01474726581211f1ad4c8f","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"160c47a62c51bb917b630c7f133d21bec932323a84cc6a13998dba298a27888b","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a0d9ef891426920a1f3ed909503ea2591418cb5e784e86f9c3da53cb0b331fe4","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2480658","recipe_identity":"f2b83f54f951a4da8b44a9dc4be85d6cda32f886c9038286d8cc2969e6545f83","loss":3.50444245450166,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"e7563731d4fb1d63ef3e7819196c82adb5cea5dad928baefaa1a4131846889a5","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"e718b043ec70c980ee8a99fd0aaf8c26b38482d931aeaddfa5d00999044b1a56","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"339cebbd33ca80f66eebde2def5fb3316cfb368b441d5669459e03a01c2ca0a8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"c420e08a6e0a8332a768b8ffcdcbfd816f0013daa79394e301a651c0c5370ca0","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"34a73ec469cea7d50b66ddc11fa1f96450d4ac03351eeb88468812a688e6f94c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"7c89f68be54867c76f79b0c4f0423283cbc3bbcfcc41a7132aead3b773eaa31a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"20d6aef3aea0b80bcbb154915df353a693b1c282078fd1d9594204d1e76cbb14","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"3979b585c7b6fd12c9141a254bcd8af5a6467547ffdfd3772e8714b828bb7395","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"8ea2ceb4d93f44dc875c0174d18b849790e43b9b51fe8f389b2ecad08bbdfcbf","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"506e02675b805243727792963d3dd6130e9a4a20bc634b0af021066857a9b2f3","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"231490dfdb18dc72e2d6abbfa93d553fa58667bad5e792f483ddc4b6c786549e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"7797009be147d6f28ecfce09c7d64a3e2daf78f9f5815cb22007d89f6367d47d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"d04db7389383a4dd388f8ffd75d1570e3c1136a82d7a73c5421de51370560909","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"23d3f35d07029cd0f94fa85dccf30832e0e8abb95216f4ec7e80895a8ac91972","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"86e600c781b99a0bcb01501df9a595c3a7a5f778aa8a47cc34d2d67febfb0079","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"81c8ca1e0862b62a24e5f04fbe7c91fdc7160fc59bba4a6d471e95f224b81601","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"aec795fa31069d34f1da5b82ead8e34c054804b1cdeb3795891d5d0981aa17fa","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"fa8dda9849919660df26d57ce3d8102182f4cc57b902b7980798442d319596de","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"b32a0bf3ebcd963421abf0d79ade18153e4ca35a0cb69e43c81fc6cc77e1b51b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"9efa24e7889f29f3d3a58da61a651d0fdb573278a1cb4a9e3a5b8c7926a3505d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"1f2ea07a4d174051bbad57630d799d7b53220d9965082584ebb80df1feb5f7ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"c8368c96fa9d89da2af9ca18418aa7f92a5ff3a0855333f9c5de2534b5c88b12","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"4445727c3a944a1d923229a700b5e325566720c56832344b09c09fde11f4be2e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"e7b8ee5c03145ccf990ec2e90b1bb74811d9f5e976cb35ac5dd0ac50e54fb42f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"a03cf4b7743b6e40f30e7a4b5ae8def181bbfb4c59dd76b09a4fddb8f1c70675","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"fede7d6dc7639ead9d1da21be6bef7c3f73f5f70098378bdb2522e2ff1695d03","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"78eee163cac91c5c9538abc28118b2e9a23958b4a1e17516cdce437c8289938a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"164004d5c62df2cc92240b50d08af9a70265f9aac864cf27a74f8eab4bd5a594","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"fc45ddba48c621fdcb770159f95d4b3d4b00c68ca547c48d3a8ece4217381f3f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"6b1058158649d4dc8bc455d3c75d8244059171cbb430533207343343159695e3","bytes":5341}},"requestSha256":"4e3f62cea6c35dd625e460fa21e0cbf1e6a1aa03e0f86bee492cb9a74c6d8549","metricsSha256":"7d03695ce03394f2ae13068b02e85281013c882b2a981e48f22f46668d76d555","continuationSha256":"8ad9e6fd837599c7221958e62a835f70d24d9a61b21eb4e2a9c1062969310bf0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"1f2ea07a4d174051bbad57630d799d7b53220d9965082584ebb80df1feb5f7ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"c8368c96fa9d89da2af9ca18418aa7f92a5ff3a0855333f9c5de2534b5c88b12","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"4445727c3a944a1d923229a700b5e325566720c56832344b09c09fde11f4be2e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"e7b8ee5c03145ccf990ec2e90b1bb74811d9f5e976cb35ac5dd0ac50e54fb42f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"a03cf4b7743b6e40f30e7a4b5ae8def181bbfb4c59dd76b09a4fddb8f1c70675","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"fede7d6dc7639ead9d1da21be6bef7c3f73f5f70098378bdb2522e2ff1695d03","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"78eee163cac91c5c9538abc28118b2e9a23958b4a1e17516cdce437c8289938a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"164004d5c62df2cc92240b50d08af9a70265f9aac864cf27a74f8eab4bd5a594","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"fc45ddba48c621fdcb770159f95d4b3d4b00c68ca547c48d3a8ece4217381f3f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"4d18b45dc92c86949005e52508b5c9b63e960a557f8ff2f61f6337524251ca93","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"990409b96ab3f9c2730ad7907396b27ee3b5727f8873c5a1f594e8fd7cba28f1","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f2b83f54f951a4da8b44a9dc4be85d6cda32f886c9038286d8cc2969e6545f83","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2480659","recipe_identity":"29ce3b688613d52c5a5c02076ab29b18b563223abe3993b727cb048382f37020","loss":3.531035744998508,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"c90bab9c6fa5cc5566c58f276e272d5cc4d6ec97ca8532866d8679a771f1a129","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"80e0b682735f47c0c0b766a45f754086fe50407ea21853160e68e9ddb3b24542","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"e1c815fb68d8bb8c0683d164110c60f1975b6ccff3cd95b46ec55388a6456ea7","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"6f4ed4c0cec1fe30eb20a85b48066896143cfbf8da2abfe051b71abdfdffde8c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"5de68cbd883e7168a25c3ac73dd89b0cf9f0db0876afa26759b2d6c02d6449eb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"192e27e48c96d12ff7ac816e9ab2e7eb41efdfc0bc7f25db3b126bb8f48fc577","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"623a5d8a53d4e41ac233f7edf8c1f0c94ef19cda811102e1da8815a66e0a1ee9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"aeeac71456dff3f293a04862ad8750b73da5f8fa20b0744791c5abc56c472bbf","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"f7f2903c938f7c2855d1dcb7b822efbdc74c53389ce02f1ebecbb5372338bac1","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"becee8dae6f59ed8a533a15aff7ba39f3540d4da875d14dd56bda1480bf70b99","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"eef27505312b8dee87ade52152b23938589b29eb5db62a5cf9cec821b2f38dd9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"6fce26ca4e5081aebad2233cfe58c6e8217107903aa9a74832cbaef2c8e9a8c7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"c25fdcd7ae3d9ddb30f4db3b313076c899b0dc8a3c419ce8beb13b25000ed85e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"ea5f1b13d5108bad873593dddf65dc8c51d71cb4c99fad399d96afa9d1043213","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"b0d5bd045edd54ef0c0309f2542ee40af992360bbb3cf595f06676c22f864c4f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"62f2dc2f6b271df3a2986f0ec6be68f18d43f82fa1b3e1b275d113805bb7cc9e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"c95eb863abbe8e74796eb53235d97f1decb0dd152d7029a7993df83fd9da3c8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"0265924834470eb810b7d1d48463ab202789561bdcd2af3ad8c594b0c0534e94","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"b56a8b6dec3d7fb02ff322f1336bd73219384a9ab83344995023230ad1e46d40","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"28b67f43919b9eaa85ab0472c48e78f69d75254151eb15548cc8a321ee01c4de","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"29d7655d307f3a6ddbb3cb096b65ca8b3094a8223de0cc88b1b6b615554e9a7f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"88a7db7d11a2058a4a52b55cb4b1ad32e4361017b0bd3cb02cf4513f508823e2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"ac2230671b114c39afe4265b0ad3f81194e4f4caa6274037ffab5070006dcaed","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"731c97bd0eeff8e0c8cd4d6f30b6bba5e0d07cb95eed314cad00ad3f905468a1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"e46415297ad936f7280b44f16bf285f029a526b519351abbea8200832a08ac24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"64aa4012deefc21cc09d9f5063d2e856f3111fbd479069576eb73cde652e47ba","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"b312ffe811cd6ebe68f04f59e58a28751384f15301d6e35ad8a8bf0464efec0a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"43e7aa7290261e8fca47a98641344a1d25174ffd4a249880faa45cf5c87d38d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"88664853f8866ffa3b8da328f5dbb7bc2429aafac2904a111c70b8d33312edd9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"792c2a975e317e6079f1e74c41582c53166a5dbe38f8f50fe2e5854f06baf050","bytes":5255}},"requestSha256":"3f86883a8f2224f839a452aabd782151c6b11b2fdf7d56700f9074579caef093","metricsSha256":"f10559093206d3a7d04a5ea9109d640cfabebf2d7db4a5efff0e374d4c337bed","continuationSha256":"8fe25f7e349048fdfc61c4c9cfee86afa4f8e1ef1ac342f740c6630dccb5d7b6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"29d7655d307f3a6ddbb3cb096b65ca8b3094a8223de0cc88b1b6b615554e9a7f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"88a7db7d11a2058a4a52b55cb4b1ad32e4361017b0bd3cb02cf4513f508823e2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"ac2230671b114c39afe4265b0ad3f81194e4f4caa6274037ffab5070006dcaed","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"731c97bd0eeff8e0c8cd4d6f30b6bba5e0d07cb95eed314cad00ad3f905468a1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"e46415297ad936f7280b44f16bf285f029a526b519351abbea8200832a08ac24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"64aa4012deefc21cc09d9f5063d2e856f3111fbd479069576eb73cde652e47ba","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"b312ffe811cd6ebe68f04f59e58a28751384f15301d6e35ad8a8bf0464efec0a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"43e7aa7290261e8fca47a98641344a1d25174ffd4a249880faa45cf5c87d38d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"88664853f8866ffa3b8da328f5dbb7bc2429aafac2904a111c70b8d33312edd9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"a219afefffbeaaaa9047abce68bff4efa7c0328893ce004f0dbaceb728d8c2fb","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"28f8c4d8ba90b0b28e2122924b1062e1e676d7ebef74618b1a1219c4be59ec01","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"29ce3b688613d52c5a5c02076ab29b18b563223abe3993b727cb048382f37020","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2480660","recipe_identity":"ca2cb3489a93c46ca47fc74ead8a62e12a62a786492db842351b464a5902e599","loss":3.526083773394841,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"a6c0dbf8b6d9407f28490850bd11c7fc6b4e2ee0a533e4858f8e4c1626aaa1c0","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"d3e94e54a5af200a3239d01fb01239d7a3e01a849bede7bf1bda3b31d45c9033","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"865f12bd9a0ae812cf1d945b6ba5c9e39aa05128262c9b70a5a1829c97c5d1ac","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"6a211bc4999aea32523c105b70ba3084f12ec9e2ecd6a848a78a0476f133a8e8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"1022ec7db56791c4d3114394b8636184db3d6919a12da50d87e82260a8030267","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"abfa6da79e14966c68b9bbc97cc5d6ed30efef7c0071cd73591fe288c56ac8c2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"009313f64de4434aad391742d5892871a3af6cc1b89666c8d68c6287fc306d28","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"96121c06c2ee616e2431e4398f8a12e6f726999109fb581a4f594be2ef67d04b","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"d9230b5f59c8b48c0fc6ab09a0ba464de132caf5a95ac93019d9961b87c34ff0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"ad293409030d8afd05f4f3cad334af244417ea37b4e73181948b491eeeca22bf","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"9d8e03acecb4def9e4824711d123c146e23b1e612f8c5b8c4686659cbdbf62a5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"e569075f0154ee44d0a8e8093ce74471529046958c7c7d916823dc584048c381","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"7c3664eb26d98d7262bdc753fed055716d215c60f200dada914b1c97504df847","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"45f578ff1886a01356522fbc8e757bdf5ad9248259a6b93813501cebf5b0bb45","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"3b47ce762ee36e4f030d2069ccf4e154a2ce0a11f79c8c6015f027dc29523ca6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"7506113143748b9ddd90019c544ec01ffc573cac329944614c48eb5cc31aa702","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"f05dae5041bb630b68a068c062b7b95def1d10260308511b4d40e1da65ac728b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"56c6573e5c0fc2c97ed3e4fb34ead140926fbb8374c29faf445049f6ea8a7764","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"63df0e1a0aca639c08a9a257b21c5361f4d606f157a8ec52fb4fb6747273647c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"12a41d8487121f520ad5c2fd5aac5efe2b33dc09e059ef3d574a84de34109abf","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"a7c59986a4aeeca12809303ecc3c6b0ffd782a6af80e885071092d57c87d4a7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"5f71d8ad5dcc7be6b79a288f6140fb57311951e89992dafef45555155f78a6c2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"58335bd8d1d70a80382e4d1d5f3411247fbf90f9287dee6218fe7cd2617b4d42","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"603ff344d4ffba1b265e6cc2a61dbecae4fc177246613e910bb21012ac50e583","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"c965d4289197b6b2e32557b6940e2341a429dddbb73620377c9ccdb0f96f4c6d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"606104c09cb7dc6c6021a24a7434747aac19b7982834de8691bb0f2db0aeb2a7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"ab7e1626d059e3cbf99fbae6e7af030df011648b82da8f3a8a1db13a1c6bfbdd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"98c22bb5ab5224a5351d28295d177d2d1c3117db6f48725f4465ce09f3218b05","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"a2d01505900d168ce59b4a2a295b5f4776b9c0de4f6548664708da37d9a13fe6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"ac0c34ad57a08cd3ac267f51d78ffa4a99354518e93640a6710bfdf5578964ca","bytes":5312}},"requestSha256":"4f050aa72c49a01e8745927deb4d50a2c3336d5060cf38371c8c02d8979bbef1","metricsSha256":"639bd2a1e74548cf136dbb88c02de217b6f5c01abd46ac5d1f7a644b7c197fb6","continuationSha256":"44859bb984161ed25c11b92d7e3bd986bd32931a945075bbea8da9908cca4f7b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"a7c59986a4aeeca12809303ecc3c6b0ffd782a6af80e885071092d57c87d4a7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"5f71d8ad5dcc7be6b79a288f6140fb57311951e89992dafef45555155f78a6c2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"58335bd8d1d70a80382e4d1d5f3411247fbf90f9287dee6218fe7cd2617b4d42","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"603ff344d4ffba1b265e6cc2a61dbecae4fc177246613e910bb21012ac50e583","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"c965d4289197b6b2e32557b6940e2341a429dddbb73620377c9ccdb0f96f4c6d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"606104c09cb7dc6c6021a24a7434747aac19b7982834de8691bb0f2db0aeb2a7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"ab7e1626d059e3cbf99fbae6e7af030df011648b82da8f3a8a1db13a1c6bfbdd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"98c22bb5ab5224a5351d28295d177d2d1c3117db6f48725f4465ce09f3218b05","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"a2d01505900d168ce59b4a2a295b5f4776b9c0de4f6548664708da37d9a13fe6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"b47ba048231ee3086e43ea2e5e794e1adf393797f1bf5ad0d1bb432f466c918b","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4b50df550a7f6bac0c9d74c43fee721ad30ed0ad14c9cb3a12e6b5b68d017960","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ca2cb3489a93c46ca47fc74ead8a62e12a62a786492db842351b464a5902e599","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2480661","recipe_identity":"afb44ca80dc09a20f57270b7ff4e65a83385214f3b90f67c75a098b6c7ceeca5","loss":3.528276419588812,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"9d89eb3cbf8e4f452fd069e1523de3e51cfecc26a9ece2bcc080472cbd6f73cf","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"dc7f61e0b36906c77f5c2a696317cec2ee1e0acb40a2805cbed8e6133e61cbfb","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"a42f5773b6eabef7f6f9e5b0f8cab79494dc60b65afcd211730b829a8b74cf82","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"92128e6ce4714336ca862f7f727f3d04e9d48c9abe2307099cf9cb880cb61c88","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"b411171ad29a0fc78fbc4de4017526966964d4ad1d7dbc1bea2bc0b1f0b92087","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"9d0148c40ea51df4b4e97a589936aae39ec2648a17429fbcc3a57da3d4c16bbc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"0f3913598bc2a34adc150715b4704b0c562dd813f2b6c819b0b894c121d87793","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"c64ba80dd7f43c1407cec72d4f926e9dedf81ae11e5bdc4d4cc25649ad54f2ff","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"f46e411077c5d3067edf49a6c2296b190240b2f5d5af0dae9cf899ac56c665b3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"b1766fd767cdc70db061637d4aaafde865b415080b6d67b88ef16aaf02333a60","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"3ee7d8cf741a93e113aa1a904d0f98943d56fed5b488ffdd8792459dfaa9efc4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"687ee7e4988d72a59ca5d78a85fc9433e19c59123eb41b26028268b755a99c52","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"0f929310b513cc9ea4b6e1f1d014b827297176fa75273e311972be1310fec6ba","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"1c4c83a7b95b05925fbb984ae93ceb2140df958165152ccfccafe011339395ae","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"e5ed023bb39bbd1d06f92a6fa899abd171e550aa7c827b1c9dfdf0ebe4a0a199","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"0935742c8ac2ad2184c3be8c9b2481211a5b8ca1f4d03b720b60eef036c05e68","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"caa2c7bdf3b876504f354b338c010cd41caf87f22a57b5e97a057f3cfef4c485","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"1d1701f1ccb8a525d3240c8cca49d387ea51227e16c56399b6d5ce4e54605ca7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"3f83fed38a5cd101e3bdf2b65ff1c5209cc0098940acc94d08debbe1c697cd06","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"8a7be783aa8426a7f1e0a106ea163dbd720f0aca03db558f930ab65d48209c9f","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"25e3d12d0b401f8cf8edcf204d080561618002c2f09d021782f757aced566493","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"3a4f19c6d473499048c4019e20d160a8407f01f5b835abcc539b974f58109e68","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"511cd78121804a590ea0fd3a72830f3dd47e2248ca9d42e0077f819245c57556","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"da6a9388830a919ca4e6dc9b852d56d2b298cd63feff1937caec8e22c0c6f1c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"e410d784c0f46aa63c930a1beb16b98a3fc9d11c4df5959eee1e8315c59b0602","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"99f20861bf9e4b86345d482c2a0b823d891ac599dd6dd9c34322afd72190b13f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"167b79183fef47c0c760a811b93876514263da6532dbd0de3ae63f9306ae8c12","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"cdceb320e24ae7efbbb7d1b2dd36776980f2aa3ad6798f6e1f64e647133df004","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"9c0ef3880d9999ae89610f437aa48464e977c9d75fba25c9dd22cac7620b2a41","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"5914b9b0a061b742c9e82d0ab1f549b0930792f041d698d358fc8476a2e751af","bytes":5338}},"requestSha256":"9bfdd0b128cbce3acd57718720092397953dfecfeeae2d276b31b2d060230d78","metricsSha256":"d94832c58640d66177e4300868b1b4f397ee6ef8f732b1baa931c486b0546bd2","continuationSha256":"ca7c7a160e443269d220fa46af2d2fc7db394b0f5825a397b0046e3d72700c1d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"25e3d12d0b401f8cf8edcf204d080561618002c2f09d021782f757aced566493","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"3a4f19c6d473499048c4019e20d160a8407f01f5b835abcc539b974f58109e68","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"511cd78121804a590ea0fd3a72830f3dd47e2248ca9d42e0077f819245c57556","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"da6a9388830a919ca4e6dc9b852d56d2b298cd63feff1937caec8e22c0c6f1c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"e410d784c0f46aa63c930a1beb16b98a3fc9d11c4df5959eee1e8315c59b0602","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"99f20861bf9e4b86345d482c2a0b823d891ac599dd6dd9c34322afd72190b13f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"167b79183fef47c0c760a811b93876514263da6532dbd0de3ae63f9306ae8c12","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"cdceb320e24ae7efbbb7d1b2dd36776980f2aa3ad6798f6e1f64e647133df004","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"9c0ef3880d9999ae89610f437aa48464e977c9d75fba25c9dd22cac7620b2a41","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"4a1ce278732aae87ca1036d01ced67cc10cf90c631b7b4837eba33573e698ae4","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"33c4376b47d94666e35d5a454a932ef1f8d73199f984b3fa26591e2f236772a1","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"afb44ca80dc09a20f57270b7ff4e65a83385214f3b90f67c75a098b6c7ceeca5","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s42","horizon":40,"attempt":1,"job_id":"2480662","recipe_identity":"20c76679f4b2d44ca5bd95fba39d9aee80845441079045c5299f7df4011baa13","loss":3.56617539654266,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"57d5cdd78084df859959a2eb13c65457481a461183ea2a1786fcbea55cee3ad6","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"646937c59ed9f7ca3cc65350c59a741ab0f55494793d8b994fbf80f3996cc58e","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"b463031e9da9ff119d416afa6853ecd65b142e25c3940e6e62d7e53d5e4ed86d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"9fa6fb094c9ed081588482654e6fc185eb0500544750072bbe8422d33ef2494b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"0b435cead175df5c71db19e43a710e14824bf2e4ee5c5ef79422c634687afc62","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"8d93a24b720e3ff7e24feeab3187104464b8de434ba0b99a4c1926791cfda305","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"8b384886a95233f74d0a3f5a689c2bebdbd08ca69d514d022570b978300e80b9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"30731b3da01205f331590aa5fa53c9486b6293a4fc0981fc7c6cc4e3ea12e196","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"8cb4ed22f967be13b17f4266a8db38804b5772727d7b72a3a58176c02d0b3bb1","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"39050ceb4ebd5f3248faf0b07e1f92336ff62f7714be4606436962693f032cd2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"bc311c77708b62bedbe399ffcc33bf057e488c4cd468d1ebd816a3a2c3e9af71","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"e45d59bf25c9058d701b9de9a19888e2575bf2befc9bf5b9a7d44f75cdd083a3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"b95257ed0af9b7b61217ee44f8ad78730c681ce479557485ae96cccc491b5642","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"82a19d5b96351792bfa919b8677868cc110afa0c4aba70f5448af704d77c892d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"7b0107a1cd5420213f4df7c0de83926eeb801e6732d7d5efa27bcf5a7502292b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"3116023035836e1c349c3bb0f96314c8647191f27c0bf10de8e9134dbc7d2ea3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"af59211c190d0e1951a01cf9492c9eee502aa10ed438d5b0fd46547d17962f43","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"3b1f734640616e701b884c181799fc44cb30b22daa5bada2dbc4c4d9ab9e9875","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"2349ff1484fd492ebf664dbb1ea8b80277ab9fe802f170c0aa5bfb9dc83bc5b3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"f3bd23c3ccdd65fc79d8d98ef1c24708dccb1707f071f76c898ace4d3bbc9bcd","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"3fbabef8dc649b493342cda1a7b4f388163f0b366d5a850b98d209838f4614b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"7f6ca4296eef8933d76a1ba537efa0043d643c3ae92c67514a44ad6f8469ad18","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"2a016946f99db3c341a3e9d6ed8554aa53f6e5d98dfe982bd06ce76ac6f67e33","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"d7f44daa59efc8df090b6de09bd342b1a5b9a77074bc1922e78eeb834a3c952f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"bec891ab29abd0444e7ae41660c1316b85b8701f7424b462b6ca2145c565f7fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"34890751a3dc4a3775555218237335ebe119d6516a5a8b7e2f8297e15b1477b6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"ced2a25d1c7807d8450c244d24ea253d1e92ea1bccbf6207bd0117f7f10e250f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"92ed6711fab9367aaa316af8da1d25a794e9c42952e6d4e37ed95db9436e0a4f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"0e7efddc4505a3b28f7a052be1d39b0005a7f5cfe5382a756123cf1e10c48ba7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"ab324b9b7a24f4b4047b30bb964e3bf8a196717ecb5c432e545b9a1943b8c7c7","bytes":5338}},"requestSha256":"213ff6d2eb5f83b9015e19e33eb4fdce2d9dd359f8cf4792a972d673adbaaf56","metricsSha256":"37630f0eae4cbd3c1bedf281d0d12ad94d17d578b03d50b7dae995459af99fc5","continuationSha256":"b9ece0a2e8f8615d27a5d1b9a0a9ca3b8ba91f00f3a3e71163a59c17d333af2b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"3fbabef8dc649b493342cda1a7b4f388163f0b366d5a850b98d209838f4614b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"7f6ca4296eef8933d76a1ba537efa0043d643c3ae92c67514a44ad6f8469ad18","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"2a016946f99db3c341a3e9d6ed8554aa53f6e5d98dfe982bd06ce76ac6f67e33","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"d7f44daa59efc8df090b6de09bd342b1a5b9a77074bc1922e78eeb834a3c952f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"bec891ab29abd0444e7ae41660c1316b85b8701f7424b462b6ca2145c565f7fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"34890751a3dc4a3775555218237335ebe119d6516a5a8b7e2f8297e15b1477b6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"ced2a25d1c7807d8450c244d24ea253d1e92ea1bccbf6207bd0117f7f10e250f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"92ed6711fab9367aaa316af8da1d25a794e9c42952e6d4e37ed95db9436e0a4f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"0e7efddc4505a3b28f7a052be1d39b0005a7f5cfe5382a756123cf1e10c48ba7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","sha256":"05b46fcb8096d1b7d25550be6f03dfc182f67ae90281c4409c91f683c4c00ba7","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0d7f5f55f924dafac11872c31147447743b41ef634edf94c7d3e5388c3736b69","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"20c76679f4b2d44ca5bd95fba39d9aee80845441079045c5299f7df4011baa13","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s43","horizon":40,"attempt":1,"job_id":"2480663","recipe_identity":"78471eb74e19389b4ed65846ca13504be2ef89a61344ed670c7c67c3cbaafa80","loss":3.5698809137446466,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"c867bd75e9ba9f183c5e602e581a65480b6c554216c5377ed7938de17d058d59","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"3d9ba6415bd42eb1a360aaab0ccd600345a879c1adb77a5579c08e0c02ce55ac","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"8cff15e5c8f4dc2bea6624c51eb1f6e0fd8ada6e8902be3b10e0bee8df80d115","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"1a3c468c56b0c7140bce6bca9c5103c13f2dc3d01bb18c458deda12582cd80e7","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"0717c2c72cfabc39dd440e41614ac01d961aabe096aefd3b81cfa12aee93eea4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"73295f5652b79d1566a8db8ffdbe4b58c28f18cc9a640b672053fe140c4d798a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"327656b86b94315a803e3405c8054b9d209af231bd45ef0d899ae814191cb1e7","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"f3c5d2a014ae9044f26f7eeb827391dabc4e85e842559b21eeb83c42a3ba454e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"36b02b2fd15b8fdf66179840c5145f17b4914ca22c1590d436b7a8045793472a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"0926c979f0713210a0fac53beda9c0e71fd56660b4d744cbccfb909054bca5f2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"8d9a1f64ca2cb46c26a7ae392384da79a6eb44b82a88849a89e646a25b57cfad","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"59bab6309eb654b78b2e87b55585869ae1b462916ade568ab974ead6d529b9fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"e3109e5d8d31f31507aefb5dd53a8f6ae000adc4e7eb0f3963b8354c77b4dc2c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"31f79b9c58f8c061796fde2f2ba82ba0a4f62e01e8c43bfab314b8ef51c3268f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"0bb80e28e00923934631b4021787417791d6b5a1698a217af4d9edc30d526754","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"ac1629a3b7ad084bc4ed9a434ba89bb4fc24c3a0fa11adfd24268fcbd0c49a20","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"c01f1634fe5acea342498e39c22966fb24879a5c1e7e6f353fbf28ac9c7efc98","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"ad9a905b0d409d5c74de265d7ff7a01e758f93e4ea7b66a07cee7a6fd43e52fa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"a309e8099a8f89b27b1db55fe5b03edb23e5192ebd660381d9662ea0faf9f05e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"5af606e9fc065f070b1fffeb43d916c8e2d82ade53b1c4adb72cc85c73f005e0","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"374ddf274c59278d34a2899b673e0d505a8a73d3cf2f0699ccf2bf454e972a40","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"38bd924034d9b15e4430324e363a4d79de96a28ef7570adb3dabd63b12a6006b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"22d78a9621b7548b78e029a4788c125ce8b3ec658940e8e441d931d070268e7b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"581eb5d6001a694609499c30f4612daca7a7a14e516c7577212fc7605370238b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"2a4ce405b7e047a944f18f14794fcb4bca9d89ceedfd8d414f2f4efd0d8b92f6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"fcc3862d05608ccc252c729865d43fae28b6b7e01debf14e4f17e4e3dae8b785","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"9a1a2d2ca4ad3db580c6f59e7a0f730654b7ec602808aef66ddab2cd6e12b4ba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"65981acdb8f10395994ac5f91ae4d5a045ce46c9f05b359df9e4221e3112ed73","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"fe7acc170e5fc16620f9113281ef1cffb2799c6f0f88a78f0d81b917bfdca880","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"d58421b8c157aa8bf44dde51fecd83179c3885c6a66acb993bd81a1f7b35e8c7","bytes":5255}},"requestSha256":"179d9950605e4c163d161c6fd3750fbf55197b1dd3c27c65027262643d3ce882","metricsSha256":"55901b7161b19bffaacc35df1ee55a9fd41d826a8d8ca5f73700db03a3b3f3f9","continuationSha256":"7fbfa75d478fca08a464bfa81ed4ad5d7d3ae4ed6aa2b8f8f80859a26faa227d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"374ddf274c59278d34a2899b673e0d505a8a73d3cf2f0699ccf2bf454e972a40","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"38bd924034d9b15e4430324e363a4d79de96a28ef7570adb3dabd63b12a6006b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"22d78a9621b7548b78e029a4788c125ce8b3ec658940e8e441d931d070268e7b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"581eb5d6001a694609499c30f4612daca7a7a14e516c7577212fc7605370238b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"2a4ce405b7e047a944f18f14794fcb4bca9d89ceedfd8d414f2f4efd0d8b92f6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"fcc3862d05608ccc252c729865d43fae28b6b7e01debf14e4f17e4e3dae8b785","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"9a1a2d2ca4ad3db580c6f59e7a0f730654b7ec602808aef66ddab2cd6e12b4ba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"65981acdb8f10395994ac5f91ae4d5a045ce46c9f05b359df9e4221e3112ed73","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"fe7acc170e5fc16620f9113281ef1cffb2799c6f0f88a78f0d81b917bfdca880","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","sha256":"aba8d8ebb1ea62a49f481797e95b2de6758deec3c6518ccb39a9f3a3a19d3c27","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9dc0b6e9e3012ec3ed0a1ebba8a81e1e83e9f9eed143f7b7bab89edcdcf2bec4","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"78471eb74e19389b4ed65846ca13504be2ef89a61344ed670c7c67c3cbaafa80","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s44","horizon":40,"attempt":1,"job_id":"2480664","recipe_identity":"3abcb8eb5b42802b8fbd9f214a14d9e4633920aeb824cde591a1e21958a66381","loss":3.5757832803340976,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"e954ddc3b77f149592b531a1f1b5342549c621a1b4823c5de54031dbc0c697fe","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"d574545df57fbd007c971f525621d394d3c8e9dd711794144d8afdc33e8ac168","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"93cc6901d8bdbdca85fc876ce40ef8d4f492cc12ad24caf47bb6198afa07d160","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"fa23a984cc279b502a2182765c342dbde31913ee40f165842166cb0a9d28d41a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"542e6a7685bb0dc62492cd4de021321cf6a6b90035d93493a3970faf10fc2491","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"f280790b5c1a4ac5c9313e04f62b5f3bdc63882e02632b563e11da974723c9a5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"23eff37b17b2be0509ddc27f3f4d573c876d382cb0646615fafd1d3ace48555e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"4397c2fcfb25740a018cb1c2583783ef0e3d7e75b019ebc236f7459e8b03371b","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"6dd7fab5996c4a9934798d53ef1e1a4960a3bee8b2fea5fe334f43ad202584f5","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"d1d7f0d7f1fb7ee89e4aeebe5b2d04419e8fa00c90ddf4c9849cb6906412e7b6","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"00e44ce550a4ded7f0994ea43695d33fd5f615155a32ed29afd8e3878287578a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"e3654df1ea4347b64c50fe46936d56ce428498d005c860f00ad244a282ffce7a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"d9ce4eec2bb93d0de875a07df025399ab7032f3204adb711ba36b1569a775ccd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"be4e7b8113b58a68337e83428c00b5fb82ad686029d7c5e99adc370fe1983ce2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"6db1716b7c5eaf35e521f229ae148bd29c2d66259325926f7bf6b7490b50283a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"e9f10e409a9a55f1cfb1631bb5507d83c3e45da4d77204312abd4b925ca5a795","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"12ff3e88fc214d971bd2766e5899a86a0e8d032be59da312c55ae5c6a2082fa3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"3b8d830183853c359375f896664d8c355bd5e75a20d82e3c9f8adec27575e001","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"d61c951313dde5b8b2ddc2e193d587fdb6d4c0c350f89719839f08de43b497d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"0d1374c6b2690a40fe2d870201993e51d555fc4b49e4cbc72573fb9d1f6e5d6e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"e7bcc78f31a56a4922c35f35617466b50a97a2d1627213242db67d39e37b58ac","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"723f672f73349c87721985a0f6e61c803a2db0488f8b54fe91fa08ea1d5e27fd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"34ac52f0ed7dc6417bca87894a8a648c7f42ed11fc0c081c07eeffbc8703c69d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"f93b05d0d1f770266c4514e3ceb7b8cc92f25e842092086eac1303c4b3e4d8a4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"c46ffbf80a19fc7c8a13e1fbb54945bef4747c4cd818dd3a2edc450fdfc90389","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"a586addd30c26c09c35699a20c4c908fff7b23ad7554102747f0a189b5e88d2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"93a8a3f1534c2d8e9843b16c5cd16aa838291db96c4f0dd720bdb8acb0221181","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"08972b96801d60a4549645d607062a9e298b1f771fe5b12d02658d65575fb773","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"54a6b85a53e72c7d3c17782bafb250e76ea341800dd3827051f167fb1c91d435","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"1ef60ce83a2d85510107ccb1c2d77d69216e517cdd132ba711bf8b8164698774","bytes":5341}},"requestSha256":"7a498122e963f7c0d1d375442493c712d072ca7ea9e0ebffa49f345bd3642c24","metricsSha256":"00976f05fd75ac5403bf2b298eae4af7e6ce02a41674e4a754e8424c2740f067","continuationSha256":"a004a7c0667a48fc819142cc9ab328e747b3fb46a7e038a49f649b80e1a0530f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"e7bcc78f31a56a4922c35f35617466b50a97a2d1627213242db67d39e37b58ac","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"723f672f73349c87721985a0f6e61c803a2db0488f8b54fe91fa08ea1d5e27fd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"34ac52f0ed7dc6417bca87894a8a648c7f42ed11fc0c081c07eeffbc8703c69d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"f93b05d0d1f770266c4514e3ceb7b8cc92f25e842092086eac1303c4b3e4d8a4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"c46ffbf80a19fc7c8a13e1fbb54945bef4747c4cd818dd3a2edc450fdfc90389","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"a586addd30c26c09c35699a20c4c908fff7b23ad7554102747f0a189b5e88d2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"93a8a3f1534c2d8e9843b16c5cd16aa838291db96c4f0dd720bdb8acb0221181","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"08972b96801d60a4549645d607062a9e298b1f771fe5b12d02658d65575fb773","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"54a6b85a53e72c7d3c17782bafb250e76ea341800dd3827051f167fb1c91d435","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","sha256":"b174f41b682f8e704c14ec79d4180ec8bf0471562a7361584df6499af91e0e1e","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0bf305808a5cb5b2424ac98b9e8a74f0e921ff024f4e1c706cc87021f7d48fcc","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3abcb8eb5b42802b8fbd9f214a14d9e4633920aeb824cde591a1e21958a66381","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2480665","recipe_identity":"c62f1e9c3aa5c978aa3f4b969c5481a5cc3e1c89fb29d640991b936d1cfff9b4","loss":3.5598033404702436,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"2165644202d7d29692e80481d6b6a6f4ae07e82ab34819f594d02ee90cd8b36b","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"5e26516aa40f1704978418248b51331968ed99d1a3dff402e771014d2fe4f230","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"60625cfb19687a872583a1bb2a38639181f8d8e97be69e13406f458eefbaf216","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"1f0d2ee6474640c692f1cfa7f8af55427b556287caa13cb7e8fad2104e9f1c95","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"60bdccf83a4137217274c9e9baf88052e19c72d513fdd7762e93c8e6228541bb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"9bd5ae7f8330a908e05f79714937aa2d6d83b2ebea25b445f7f2b6c9250ee68e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"989c9386373edec8eaaecec15863ecbd1d7183e3c09d377c8509361af0cb91bd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"3f8ef5dcb916366b87abbcfee5d233a4a5acd4a16f0a6f79de2ac3be16737814","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"c855877e91b968240c06b95b4c19686fef0be74c001eea8682128c294f98d0c9","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"bb3be74782f283a2abb8c223d80e89c7fda6f8ca4cce718de2439ce4d64b36ae","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"384deb135f7e48bfb1c4a2f7332b4fbf8a133cf7c7a61347454cd14b60c447f4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"fecec5b9a9e6e5a54146da904ad1987160a1980b0371dbde855bbbbc1fac7d0f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"8b1f198c7b31b4c723f122147425b3d4ef29c4782f7c96c3bcd71d1791a03085","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"24c631d257c4266b3b72258771dd1d639914a2f6135bb0dde4de53ba2c97dc76","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"2bf6190c8c0aa085d450cb2dcf0cb8e0615d5142accfd9138b7293eb1e75aead","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"04fa823295473125cf626ec8cb5cf6a89b3bb22fd38ebccc9b537ebcaa856fe5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"582093db9f4646b8aa35394eae927c1573d3d7c37218996a05eaefa3af206df4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"14350a8da92951f9aedd8384d29b534d418bb5537ba1fcd7a159aa22476ffc0e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"c6181b7ff766816e19bc865f85ed63fcf3d2127fcc9eddda972328c9a31d2560","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"da708b7d419115ffba81a28a64ab14193ff9bc74d7c0068f1d9a13e613f70999","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"1cbf195798e0e8fa577f41067b8a32c0fedb98f787d82339c618f59c0da79a80","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"26d059fdb835b9ea35e91e2c86c3755e280bc9ec18253790d7dacc5f699776ab","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"15d67569dfa6fff4791bc6f7054014c99f81c18fa71ae0adab897c565d23006a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"ae030d4161144f0e1dc7502a0022bde3fc82d47b68e3b347afc3a04bd5eb3734","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"2d1c07d2c3fa1deb5b654f952aef9c8b4dcbac1b6e850a1de323579b1bc80a7f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"32ad4984fa47ff78cf76cb42d8e70ba118fe94c7efffac744da7f663edc01421","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"fb484f085bc7ebda4d19ae4441c2796c10f57f32ef1b8c32fb81b05b220f21bb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"7259ed5a24f79752e12b48c1765459e89d53aee53bb8fc3adf91128d6ef9ff41","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"830fa65428fbd9bb73e4c0418948b04dc7e7ff3d45814e8fa732b58ae0854143","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"9a3d30664a9010ff7557f5039ea32fef2a743cca1d0228471fa29f8f9c380066","bytes":5341}},"requestSha256":"e934147c132021f996e0b71fe9796563ac1237b120659ccec66116c6ebe2312e","metricsSha256":"145f3290e2169cecf88db69496dd69cec435512dcfe35c549fb74d70dfe06df0","continuationSha256":"0e40a6bf119f42a1282ff607cd4be230a97ed7b4349da787d4880249daaa2e83","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"1cbf195798e0e8fa577f41067b8a32c0fedb98f787d82339c618f59c0da79a80","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"26d059fdb835b9ea35e91e2c86c3755e280bc9ec18253790d7dacc5f699776ab","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"15d67569dfa6fff4791bc6f7054014c99f81c18fa71ae0adab897c565d23006a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"ae030d4161144f0e1dc7502a0022bde3fc82d47b68e3b347afc3a04bd5eb3734","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"2d1c07d2c3fa1deb5b654f952aef9c8b4dcbac1b6e850a1de323579b1bc80a7f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"32ad4984fa47ff78cf76cb42d8e70ba118fe94c7efffac744da7f663edc01421","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"fb484f085bc7ebda4d19ae4441c2796c10f57f32ef1b8c32fb81b05b220f21bb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"7259ed5a24f79752e12b48c1765459e89d53aee53bb8fc3adf91128d6ef9ff41","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"830fa65428fbd9bb73e4c0418948b04dc7e7ff3d45814e8fa732b58ae0854143","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"8de3e3466cfe0fbd0714ac6052dff7a11d62d5887a48ffb9fc2dd9259201bf1b","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9b78a60c8436fd0ef5fbd3106d51ba6701583f3309db5b935cee444396966e95","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c62f1e9c3aa5c978aa3f4b969c5481a5cc3e1c89fb29d640991b936d1cfff9b4","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2480666","recipe_identity":"b201b665385ce6b61f50f6a3280d284d30af2d6c1b37d19a25221cb2082eef38","loss":3.5696122454449406,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"b5c1cf7da804768ce8a9feee4fee3d0486af9f6ba55e7bf868b3ef5a53199cf0","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"fa36cd4d758f2b51552b18200a3f8d57bf6d3b415851fda5f173d0b2cb747fd4","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"4aba3d2a85dbf2b32b37aabff9d240f30cb231795b3ebe3c7de9c9cf531228f6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"43940db0e81a40bf7ec9c43b0fd6530b2ac5b14a80aa3038b6e0cc4cbeb2d7b5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"1820fdd6c9cd1dbcf383ca26674590cff489a56d559c292bd1e56bc40107775e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"99c347c2076c70d37eaf6df5de5773ef735d5796bc7fb22f4a5bfc2b0111c449","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"46568e46122feca66c1479ea3481df5217ea98a747eea55f129af9ddb35958a6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"da83bfcd11bef6fcc1db4bf6dd1eb941cc05db156dfb82d17eb99cf9fa450e0d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"6a96e22dc911a1b1bf256068b09a750f29e26edcba62d25efecf1fb602af69f3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"4f2b9690908f34118bed37ffd1ff0ec2cdc9a20bf1c2326b1d01c5081bb0263d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"7f16e1cc43445d91e763d707c46c40442cdde418611bc7a64d86db0723becca5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"6711666ba5e3ebd81526ba9fb81a11ae2faef380a9c82bfea86d46e9b94d1b71","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"f591e9e6f06d06a7ec8fec54f33ff6325f340525ae6ba7b3a9b4debbb8182927","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"2314d9f426124caf5d47fbb62a8962bea12013ce1d36f2034b99962f3edac51c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"9364090f275965a02c664c1b0893ad7ffc7440d1b314928f5c257f4c9f77cfe1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"1b0861df0d250a4ca4b987dbc0fead7a034cbe5269596e8f14c8ffd171d35cc4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"d7bcfda9b3eafb89d233088e7829a566e353979410ac911d3b03b4e93a0a9042","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"fc9f15bf8c5d6d12bac87ca30dfa91d1f7c8391607fb39b7d2b9531cc2f59732","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"f10f43bf47a49304fe68fe0dfe4aaa693b69d395228e16436325320103a0ffa1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"07541bc765950faa5407086098e142ec26c5574b705018c7d301d38dbcd60c17","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"9161635c523461286817912746cc603a10e209657abca2adff827865ee3f541c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"a07592d1abcdd0f3f7b2592a4254a71cf238afe1dee3af31e43f2ee0aba5fb16","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"f3fbc3d682cdc7f8e25b6e18b22b105ff7b1c2a018ac17e9170aeb9e4d2c7060","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"504fb40938aec512a2489d8f86e44269205f2dbf76b0eb3e3e5ee28bed4f8bf9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"3b1dc25368678ea1e1c0438e878f715e821529eb3f99748b79830fe7dbf27045","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"9dbb735de493a23181537c0534dd29b399c788a39f7dc40671bb30e61b2adf6a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"21c048c7df3c9a21760f5cf6e1b0344848dd333ff7c942c0922b333bb210881a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"052a43b32fc2e315c7dbc1c6baa5a09ecbc5c88593d96512f87cb025adcfc8e2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"31c0a78316f8ff8633134b9eb1afb456cbdd9adf64da591fe2fe47daa3fc63cc","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"88ba9858d303a90d750ba4c8e419fec18c90a74ad8b9a9f894b1695192681aa5","bytes":5312}},"requestSha256":"f603f0b23a9da00c08b143465e53e8261df10a2d1f52a69a126a03daac9c4c4b","metricsSha256":"fa71156cc66d47e9ecf4dec0cb5922077f467f8d389f74c01f8cc6367ef833d9","continuationSha256":"d8a9e2222098ce9ddc912ab888c88463227e8bb42b2c3c3185dd6b6da2a7be7f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"9161635c523461286817912746cc603a10e209657abca2adff827865ee3f541c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"a07592d1abcdd0f3f7b2592a4254a71cf238afe1dee3af31e43f2ee0aba5fb16","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"f3fbc3d682cdc7f8e25b6e18b22b105ff7b1c2a018ac17e9170aeb9e4d2c7060","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"504fb40938aec512a2489d8f86e44269205f2dbf76b0eb3e3e5ee28bed4f8bf9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"3b1dc25368678ea1e1c0438e878f715e821529eb3f99748b79830fe7dbf27045","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"9dbb735de493a23181537c0534dd29b399c788a39f7dc40671bb30e61b2adf6a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"21c048c7df3c9a21760f5cf6e1b0344848dd333ff7c942c0922b333bb210881a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"052a43b32fc2e315c7dbc1c6baa5a09ecbc5c88593d96512f87cb025adcfc8e2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"31c0a78316f8ff8633134b9eb1afb456cbdd9adf64da591fe2fe47daa3fc63cc","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"93e67478f4e937cb3211acc242318697a6566b243a1a5d355d3dd48078c75579","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"400eb85d6d24e6ea605b857e408d6f79853cd856ae6fc343b57abf2758d03d92","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b201b665385ce6b61f50f6a3280d284d30af2d6c1b37d19a25221cb2082eef38","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2480667","recipe_identity":"05e108a2ce9345a4ae2ba41eff7a1d08bc4e76a41d7f192204706bf074b757ab","loss":3.5710064930191603,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"b2aaff7d45874a7446db49dd72f92932d85c9698202294c8e0aeb9caeaa1487b","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"b2556ad1e53f8ba24df549eecfee4c004d3d067e4715048abcf666acd2d7f718","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"d8d96155aab9173c520cc5252a913faf19273393c555d8062a9de22d4c6cdcdd","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"d1a2218a5370225c12e479b96be06cf5a929e5c0aa2fab9039d50736fec47aa4","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"c103f8c7ccb6974cbf0bb985dca799a252de1c93611dfd3e087b30570d051ac7","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"e6d64ae684e24dd22e5a4ae57c0e718c5ac0c804e8f7825bf1d45042bdbfaa2b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"410797a914172456fbcc960c9fa7e79090e380ecbfbfab8b3e41fccf3bbd11c0","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"df28bbda2437d9f6d703a10fb1f8063d4e4a1c24c1a85772747f2681e60a82fd","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"543ad3e334d7db0ff6b66aafbf4c74159b8a47ec1a30103e0ffafbd84f3b140b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"a633ec191145e46fd0dfdf90aad74d93fb99b3860f7ebedcb446d29b3335de6c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"4560d59bfb242556f565a8b7df9e91787eccf544ed958cdf6ef91de990b34f3e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"02484b11ba09fc2082d01402293e903982a3ec9c4a00a7e3e98bc3b2c6b903fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"538195540a03b59cb2ad31294eb4949e0fe71b3ed3ec6766510d0315663b765d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"c79f1db301e9d0c4e76f9867e09f6e4f050f25206c204a57e5dec9de53748d41","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"371028704a915c360652d9f610cb85afe84e3ecbd3254277008d2ed23ef9c38a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"90d7de16add224fb81f8011a9c19da78c33844f9a7b648a01598aa47640abe85","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"bf0305feb63b4a5e96128d437c5d38b7f0fdf41314ce8b3ad79942a7b9707de3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"b63687a85ad111bf8f03f54aba909c918bc8fe160669e95bde6bdaed90b5e293","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"46c3de7dcd38399eeb39d91b1cc6f3c2c20e61ef874ab07a44f8f579a8609671","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"b10df18dc4d579fb2c81d335dab0cf73d3b42ae39db2bc4dc76928d382275567","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"c279e159107777231183efb7ab8a87f15fef38c9738d6a6e23e665697b5672db","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"2e0966583fe669411538f53e2902ecc9ca3bfb64a9ec9543ee43073258ea36c8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"138f78b5c3705e0238a5e698905f3afd55ef0f7be24bd7367bb95d991a5bc2c7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"e271a21dda4d40aa666053e9e5b656b05e84b86afd91eb791db15c53ecbc73a9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"4b229c31e2e86436820d3a95e8b41dc42aa4d2f9ea06c5a8e669f87ece7e3d24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"36d6f3ac5c570e2f4e840fca1cdd28196d0e29247a6aaa787da79a3641fd891b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"2cee3ad05c7ae2fdaafbfb5a77f808e8ccf26f22217aa71b3f7d027a3644f077","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"6940d890174fc6d504ad99c5668153c3adf5173e704e46ec32299bcf88175e4a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"4c7637caa3d78090bc0a4f7e9e84366660f2a4a8b6c6ab9f3d4437b2c73a6cd5","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"c4d7045b15e9b7659ef4766b9535087cf38c65011b723e1cfe04c27f4e9d4667","bytes":5341}},"requestSha256":"cad00f96336304802567705e356803bb954f1cab4b8433a05c55267d24c85e4b","metricsSha256":"08a2a468f3b292b8b3f1ba35c8e167f9feb542bf00a03c0c0a473767cf1eba2f","continuationSha256":"f3733ef833d83a4f3292fea56581d84bd7e392cedf1d240a6cf666aff60319bb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"c279e159107777231183efb7ab8a87f15fef38c9738d6a6e23e665697b5672db","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"2e0966583fe669411538f53e2902ecc9ca3bfb64a9ec9543ee43073258ea36c8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"138f78b5c3705e0238a5e698905f3afd55ef0f7be24bd7367bb95d991a5bc2c7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"e271a21dda4d40aa666053e9e5b656b05e84b86afd91eb791db15c53ecbc73a9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"4b229c31e2e86436820d3a95e8b41dc42aa4d2f9ea06c5a8e669f87ece7e3d24","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"36d6f3ac5c570e2f4e840fca1cdd28196d0e29247a6aaa787da79a3641fd891b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"2cee3ad05c7ae2fdaafbfb5a77f808e8ccf26f22217aa71b3f7d027a3644f077","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"6940d890174fc6d504ad99c5668153c3adf5173e704e46ec32299bcf88175e4a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"4c7637caa3d78090bc0a4f7e9e84366660f2a4a8b6c6ab9f3d4437b2c73a6cd5","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"32200b73ae05300f14d6c2759330353562d5f5b37c1798f378696dc9720bb525","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d55be4bcee03c6237a0713ec344b406528cc71085f5425f2ef20a9a44c0d4a10","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"05e108a2ce9345a4ae2ba41eff7a1d08bc4e76a41d7f192204706bf074b757ab","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2480668","recipe_identity":"8d720a3e981bd355b6f1f21443093cde8b3217905f5cbce3935c8d946cc2aa06","loss":3.5714906390281143,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"273002a7cc4f6c33c437f99aa19bee31191cdb13b8d3c244d0924763fda2ad96","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"31085e9a889424fab3a3369f66c2188939eb4c2e0490d730f5733d711ef6ef04","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"540f4381e7100dae1b7c14b46092961d4e9ca8e0b224f77a14580403f479885b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"15e6a6018e6456c9f77a449f36909fa1a3b5be8dd026e7a0d921f501a05e5721","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"b8e2682f61c9b685caa64ddc0ad356fe038fc72e79635cf7638b4ceb7bcf5b12","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"d4a7d83e7ae94052ec909a59067c0ecb5f77e231680459645b6188bb992c9962","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"808a2af2f2493bf182b60625e043872df6402461ab74f49c32b9f617be29750b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"d8415ff1e6ee9141f3303ea7fb602e2d97a58f98865042b3d4ea6c2d9454baae","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"3cc0394e51d902634f4b533fa7a76bbc19bddbedfb92a56628da6786cb78946d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"d4114ba9fb06ef03a69d66911e6625fb402eb89374fe55efa1f5bb333b888a12","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"5bb1cf00ed312a3b5cf5b152faf16780700f3e55067f5f4eb807a10a773de1d9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"b06036928109d4882768e9db3a7155cf3011cc0f87c3ef0eefab0b70069743b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"2ea3b6ee13994d5c7c7451a64799dfea8aa9f4630c8f9826e50e43e9606cd88a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"20dec9f92e396c3f2289bc463781fcc0c2cd0ccabb7884d90de76343bb3c99b2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"30c82c91e16d5b8b091fe877322cacb526d6a652fefe807d36d15dcddf08398a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"da982ff5e0ef32f03e3cc4614b3f0aa7f1aacf32e04943ee222c0b823c3fcae0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"703ebf33d9bc931a8a11fe80e982427514da24409e767a5243cdf46d241b2456","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"267c76a7fac155d86eadd13209dfe642804814bd94c6cc21b9652bc967eb04af","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"c8f626778e5ce70837df65755d62fd507828ef591913a139bd03f4fb72915a4c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"e8bfaffe2c146fd46e5fcda34ca47536c677a9aa01339bd0f6924933cd969e9e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"8346ea2b483c9b4a87bc024bc4535af998bb3f83b27c91b0488f3a169260b8ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"d17d7210b5ba7a1fff66d25e46bacc28e65e61179a79cd27c46e0e1c4702009b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"39635d177c4842253aefd0d319f9a585a0a7224f974f3acbe449d24bc7b72b50","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"334b01e5d067b18a8bc188bebde53db5e07aad88056082ef24ece9a6e2188da5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"a2a14f8471a326b4e5e7b71e984a68b66b91a10e095658d2eb8ee943dd6898f9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"b8dc00aa4f7481c795f5400621adbe3537d2e580d7d98c85c8ead64ed30a065a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"b63f342f87a11f026d2ea2ecda53a55befe89ec783e825b4e7dc4825bc3088a1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"08774e2270f57dbf6fdcce909078a026c12effe1eb07e69268b1da953741a9bc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"7c3db5fa422287e79c88fa21e353093756c9b145dc5b4a31653301fa85f5e7d7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"d2c7cdf1aa48fe60126d72ea4ecdc056d6ca3966a4c23b3e5fea10b0eee2b247","bytes":5341}},"requestSha256":"080da0a3a82d843b29c50836cf596093a5a6fcb049747b96d18f52b119c949c5","metricsSha256":"ec5b7526af455def53289fc77c78ed4f83261c41d7166727722b5b4d11361682","continuationSha256":"845b165cfb96052f63b74cb2fd1ebe1dec0c7afdbb101ce25c0d78c6c97adc0b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"8346ea2b483c9b4a87bc024bc4535af998bb3f83b27c91b0488f3a169260b8ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"d17d7210b5ba7a1fff66d25e46bacc28e65e61179a79cd27c46e0e1c4702009b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"39635d177c4842253aefd0d319f9a585a0a7224f974f3acbe449d24bc7b72b50","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"334b01e5d067b18a8bc188bebde53db5e07aad88056082ef24ece9a6e2188da5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"a2a14f8471a326b4e5e7b71e984a68b66b91a10e095658d2eb8ee943dd6898f9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"b8dc00aa4f7481c795f5400621adbe3537d2e580d7d98c85c8ead64ed30a065a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"b63f342f87a11f026d2ea2ecda53a55befe89ec783e825b4e7dc4825bc3088a1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"08774e2270f57dbf6fdcce909078a026c12effe1eb07e69268b1da953741a9bc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"7c3db5fa422287e79c88fa21e353093756c9b145dc5b4a31653301fa85f5e7d7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"f88eff292340391356dc3c766c4111a314310a13f8bc0369c4b7c859eca29ee0","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e6b21d3c15eee2a62124cd752243a1ef45eb7af94e6eeb53cd7c8accfe6ca955","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8d720a3e981bd355b6f1f21443093cde8b3217905f5cbce3935c8d946cc2aa06","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2480669","recipe_identity":"a3088b8abb561806032d0f55e8cd6d6e6cd56b0566940ad7dbadd49b9551050e","loss":3.576699536579847,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"610cb01124959342bb36febe717d271ede7a319f29ca86b82185a4d65703e5ab","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"8bfbdfb91b71f47e7b4e4488c10e46fc1657c94ca24107568a3803c97abfa26b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"55e9ba3b297eca9da6d055a2bf5b6b056a0c98bde885c7c11d775d6f84456006","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"bb67af8aeda0c0eefa5803467c39352c5fdd58b065f049bc095e8966bd5ffc4a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"55f79cf3915f40f2022f61762670d967bed01cbde7355aa17d1f1d86dadc542c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"4dcd1ad99af4198c4af4ffc4223b773509bc216bac9513997b6f137e64e659af","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"e500ba309b84e693a4f547372a6f66cfb4cf3c054370e630389f232b2784c29c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"502866ead550d673745ed89038d7b0548ec16ac1acfc0758f9faf67eef32950d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"2a199ea709261dfe23480395f7771e16ed298240a34ff174c617635f5a4b4589","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"3cd87722ce21bb202fbdd1f39e08eb8f92a2627472c84412cefb21d10d25ed7d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"8236e779adfe1f7e4aa76dd0a937ab3c581aa38e57310e35a1c3e1a40e2e0fd0","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e0b7d6e394867f69a1c7f9aaad2e824414d422282daeee03c692f07004931064","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"98645167fbc84f5e1813880635bd17d5237995e5b8d57b644d4ce7dfd57818c6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"ebc0d4ef6a111d6f00d5c0e79bfdc42e7d17d34a118e179f90fa652282c9ec59","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"3e1ef8933d9918992fbf9bc7e913d6d386475f8b3bd649841beb0ca835f62f2c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"d8851a232768e6ec9e2baa3043e8f130b3f4ae21d3ca40c1101d7f21322edf4b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"096012e2e6c3ef8d82394eb327a72ccf0a1a66f0dae377abd6036606446d6e50","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"f7d0ca36dc51f2b108edb123b5359ff8f7a5d7d927564a839a693be52ac95e56","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"f8bd75dea5303d53e62996d401b03eaaffb505f797a5c78a35a9371ccfd38afe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"7c7e79dd079a603a0922df5d4b5fde0ed12ca455c0133beaf564aca8ad950bb9","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"bd76947f89a8d1300f0c14249881f8848f778eba5f3d46e6433c02fdea0379a5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"29afe611f5aa882e281869465a5af2db3861b893841710220297642a84bd2d4d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"22989395a399d392e10e54523f83d2c272bcbdc4ecd4af73636716f4e614b1c6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"90a9687ca7a89abce275fd977523ef5c9102d79dbf0a0e1d809605d03637673e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"2e044bfad7100b9fbfe97782a269bf91b45933f0bd7d8dcab81748c2d542f350","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"6843b6e458b0eedadf4d3b012f439d44c302d4b39a5d46a37cb3fa61a9348482","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"cb8c84d6c0b39b8c5769aff8459a33bd290e2680489a0c100385e257023d5f65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"ed1e72fb4a937f711ac6fbc3fd5933fd3e599f1f6a0685433c05a1b531e0b96f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"31267a65e3f92f998cf835b5987d058369b5bf9fe781559e14b4d03ccb34a592","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"f43dd42d5bfcd207ca5e27510720d385f52aa19aa2495d770c83e08716efe52d","bytes":5312}},"requestSha256":"124342439ffe0718adc8e0917c1492fe0009153e6547f189f125833c181a3c34","metricsSha256":"481e37357c70329d6e0be781ce0d95ea0edf5f229b2fb61970837180117fd84b","continuationSha256":"9cc16b13d2e5c16fbfe9ce51d5749eb7aae2d15596eb5ca8b621b8b6a912a15b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"bd76947f89a8d1300f0c14249881f8848f778eba5f3d46e6433c02fdea0379a5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"29afe611f5aa882e281869465a5af2db3861b893841710220297642a84bd2d4d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"22989395a399d392e10e54523f83d2c272bcbdc4ecd4af73636716f4e614b1c6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"90a9687ca7a89abce275fd977523ef5c9102d79dbf0a0e1d809605d03637673e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"2e044bfad7100b9fbfe97782a269bf91b45933f0bd7d8dcab81748c2d542f350","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"6843b6e458b0eedadf4d3b012f439d44c302d4b39a5d46a37cb3fa61a9348482","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"cb8c84d6c0b39b8c5769aff8459a33bd290e2680489a0c100385e257023d5f65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"ed1e72fb4a937f711ac6fbc3fd5933fd3e599f1f6a0685433c05a1b531e0b96f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"31267a65e3f92f998cf835b5987d058369b5bf9fe781559e14b4d03ccb34a592","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"304a772bf572b8e0e35c017594beb54fd2e5619b00a431d0eeb54a6cf2a50413","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"984cd34056777b030a527392f27423dd95d8a5c048cea43e319f4c5daf4b2b29","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a3088b8abb561806032d0f55e8cd6d6e6cd56b0566940ad7dbadd49b9551050e","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2480670","recipe_identity":"2a60dc08718f51d6341882510ad4d120ca781d12ee4c3c118c3d3f061bc9eb48","loss":3.578922651370372,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"e2b0c87f118a7411e295ee2160b576ffc0affe649395094531b06340b4f4ece3","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"95447de0d9dc16aa6cc320bb6aa3b40d879a750447c3ca17649fab63d2f9a152","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"fac775e7e4b51f699ceb51359e6619519f9764cfbca71be6ac26373eacd16127","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"c8b028d328a6226b4a302c13bad119edc46f7126cb83cc714102bdb241e2dd51","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"ff1502903ef79fb58b0d01eeee866067b0c91814ecdebf4c17264e581fc68774","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"833335fe3fd89bdd6769c9146b7bb08935fe813ade28f02b93cffda5d157ed30","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"8da866ca482fbb5d37b53731aa89aaf1714c9154b900ad55010979274417eb47","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"f9f762aef5cd53c04d81400faa739fbfe15a5f2737d2fa31aa1a2719f2e52563","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"4323f3a1eee53f626db76098d360e2de6fcc454676229f5c35e425fa55196cce","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"f18646f2cede1e805cb8ea45d63442acc447a3e5080d12ce176fb02768c35aab","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"c52f0eeb211539c79f58694e9751497227eb6fec8dda49ea2ef616f0a9dbe77f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e7d0de7d328d711233ac451477a9c8dbe5a709c0733f597315264b25a6cf06ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"29ce46035055b284b19c6bae8d8355ca734d3e3670f36987b2d426cb05e8d61a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"abe1567e0ec94bf21597c5cd694cd2154af7e1a1812b564bdc29fb8ed7c8dd3c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"658cefbdc3d5f49ce742d62172cc1f5812942dc5b4583aec9b6883973b71e38f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"5f2e8a926154803b800b5e9a055afafbdd95c2759a84494f94d2ddf042e0c401","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"756d6a8498dbde29284487c72461f281c84432f53f217e1bd989da9c824f4317","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"28f2246fc5281e310c51ef9850de2ee0f86ed06665dec1dda3515e37aaa75c65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"e07d2585e1b382df0761ec785cbb0c5b83557ae85bed95d2ededaf8dbbd4c27b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"c6af6edae6fa3df0136fd89b3b73b2a5323643ebabce993cc903a9bc18cabd61","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"93af5e0cddef2b30ba72738b7c9cb6cebd5d210c7811e3f73d5bb09acbad202d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"82244b7dacd2f402cb7efbecdb6e41e3fdcd841a0e9ba4427725cfa51c0e3b15","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"c3e9143c6f95b389ab834b6e668c58c16e1a71e45a24952949c1f24344ac6148","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"5f2af2981078774c84185847fc6f3d912dc99e05da3e71da0e9057ca6916cb0e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"44953598b12a9ff95b731a834fb869cd38ad768069db78ef76d79f0b98265300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"15f9a57eea8241fd6a9d6beb587fb9ed338245261cc2e5318efa6b08f5846ce5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"18458920691da003ba31dffa6e6c65f61af30aaea47f914ab82f14d3f31aaa2a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"ea1fe629ea59869e63938244922d622d28fbbbfdd1680649bd75f1ae1dd8100a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"4a8a86cb9d4b6b58a255ca9bd73c29e84d59d11bd92cd2bac0afb2b9dbe7f9fa","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"502050a32f3bb3acdad0a526572153543434595c5060ec9212e6c3348ead5bcd","bytes":5341}},"requestSha256":"534a3409894b1d604e574079b0ecf11dac9ded292cb2a066537b8249c65a7b59","metricsSha256":"a78ffa840a166f850c886be06bf34d28e9a40ab8f113531593acab1e44db1d66","continuationSha256":"bc20b1bcb67cc90e2855a1c8a947ea653efe37fb453c4110a1f896a29125c848","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"93af5e0cddef2b30ba72738b7c9cb6cebd5d210c7811e3f73d5bb09acbad202d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"82244b7dacd2f402cb7efbecdb6e41e3fdcd841a0e9ba4427725cfa51c0e3b15","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"c3e9143c6f95b389ab834b6e668c58c16e1a71e45a24952949c1f24344ac6148","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"5f2af2981078774c84185847fc6f3d912dc99e05da3e71da0e9057ca6916cb0e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"44953598b12a9ff95b731a834fb869cd38ad768069db78ef76d79f0b98265300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"15f9a57eea8241fd6a9d6beb587fb9ed338245261cc2e5318efa6b08f5846ce5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"18458920691da003ba31dffa6e6c65f61af30aaea47f914ab82f14d3f31aaa2a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"ea1fe629ea59869e63938244922d622d28fbbbfdd1680649bd75f1ae1dd8100a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"4a8a86cb9d4b6b58a255ca9bd73c29e84d59d11bd92cd2bac0afb2b9dbe7f9fa","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"0a09ebad2df5db9e32eaa9431525994de9f0db4fb315c48ae423e7164f11dfe7","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8b6e4f0bd34392e9bad8cf5928daa86ccc1295a6432d2de1d0b7afece399fa86","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2a60dc08718f51d6341882510ad4d120ca781d12ee4c3c118c3d3f061bc9eb48","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2480671","recipe_identity":"0695af5503ef72ac20eb9d69187842c955baa5331ed727f16df81c59243e5c54","loss":3.587098174154891,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"e36380802ba29ea996221bf95de2764a030e5587b278f07820c758131ff9849c","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"a95b676d3c2aabaca7127eeedc675abfebbe80437c4832e285e45a13b4cb7ee1","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"cc6438bd167f5a6cb3d58d89e0b181146e2a57a312c33953606e5b946bb101d3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"dbdf2662bf638f59f23eacaadd2bdd228a10d55a1792cc924b3b212ff695d11e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"d8ab3f93b4587f4fc728a6a4751ba1a9a4a0a596c4b1c1e7f4cb3d30699b1791","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"5b6ea5e8e013a6eae51a5ed6a38a058b2baca49b103d97a5efc57acb7590c383","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"274a795a735f354511f13124da31559963297eb6b1b3cebc76e0a9ec65946565","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"a11a1d44a367c47181a9cc3de6806f63af80b69980fa5bc1784804f4fe09e204","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"01dec99731496373c6ab0b4c17c88eefa5d6cfdb87014acb86fd7ad5ccd59001","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"300d939766dea65a573263c3c63b0839b1522a830824895511cb99e7aa6a236e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"41a2871b01bad85bd71b226ee9ded01d6eef79bcfc66ce94ca37a06d68d06235","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"1f0c2adea6a68f992f222c24fc24e09258b3a4c38fc25b2c1832007a4ddd1a7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"c22cce35d1aa5a280f49c7fbccff57c19a11550abbcaf5a015cc2ce1f1362a94","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"2a5651d3d73e59bbed2f39fb819ca738bab2b66568300775b0c54ae8fceb8c41","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"be9a1b41bb9738e6d7b77a7c346b1e21db64e9c1971353e264e87315e93bfd32","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"8ec473fb694a8f3671779f98111d83bf90b30e6f11338e48e1f93809a751d9fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"d62bbba2b31be32a8b645eec8ae9c93581a2c8a0e392295466c24db2a478b35b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"fcb7654ea02918158c53d8bfba340f95b27c649c205da9a5bb2cf2b8e607a2d0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"211988b166fe5ef30fd6ed3f5b006f54c8d69a992d809a34b17973fcbc851b9c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"bac3f2ff77c383c6d0af49a706d2eab6053f823bd8f30fb08396e4103a5ec671","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"e8e5f262a2b19fcc4edf005dfdc5254946afa0a8155b550c18fb0e7a0228c420","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"cf24786bbabd74ca74daa883f74500d696a0a65238bb3b2f8245e381eee89ad7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"f72dc37b380e58b723ac76e959c2e12d3e1f3886be713adc598c82d4cc23e80d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"e7499926597e48da2526cf03a69191c62dc5ea620d8a3f63b26786e0506a1700","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"f7bbaf386870cec2af3cf3c05d8c489f32282414a8033ece880baf29f3b8c5ba","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"a1d302970e8c8fa9c07abe02adeb4b224f02d85e48895deb8f828eeff43f5a96","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"98c18ae34b7e378e1ce1fa0f00403fd7be064b1a193f93dc959bc487301c21f7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"98b1c7857693f4054accc5a6eed74df1493eae26a17ee749aac8b911aa2ad487","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"c9e837bca46bec4688f1cf857c9b4bc5c0e94dfc1d7599392ba8105191584971","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"4d8f5cd7564a8910fc08219480c3959bf1cec558ecc59fda100ac0d4a0627fbc","bytes":5341}},"requestSha256":"fba6e6ac7339e3d5ef1ac982154d4eb3649ef270ea95e711432518406a9a2cf3","metricsSha256":"0280b4bf1b28c71774415b931a08e01a29d044b5c35f82632cb920646d72632c","continuationSha256":"668fab7f5cebea117ce29374e1d0a3d85cecc17f7f8376aa7bab6ececcf234ac","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"e8e5f262a2b19fcc4edf005dfdc5254946afa0a8155b550c18fb0e7a0228c420","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"cf24786bbabd74ca74daa883f74500d696a0a65238bb3b2f8245e381eee89ad7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"f72dc37b380e58b723ac76e959c2e12d3e1f3886be713adc598c82d4cc23e80d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"e7499926597e48da2526cf03a69191c62dc5ea620d8a3f63b26786e0506a1700","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"f7bbaf386870cec2af3cf3c05d8c489f32282414a8033ece880baf29f3b8c5ba","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"a1d302970e8c8fa9c07abe02adeb4b224f02d85e48895deb8f828eeff43f5a96","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"98c18ae34b7e378e1ce1fa0f00403fd7be064b1a193f93dc959bc487301c21f7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"98b1c7857693f4054accc5a6eed74df1493eae26a17ee749aac8b911aa2ad487","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"c9e837bca46bec4688f1cf857c9b4bc5c0e94dfc1d7599392ba8105191584971","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"e79fb6bd85d589e138a6eaac3a075f96d3e09678fc92f16fb1a355acae6afda5","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d5b46747651389a2a7f04e3ddcd101dce69b17fb70b28f58d902eef02e9b5dbf","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0695af5503ef72ac20eb9d69187842c955baa5331ed727f16df81c59243e5c54","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2480672","recipe_identity":"8dbec3e35359f2783ab9822db33ab167effb71c8414780394a77534afe98605b","loss":3.5880995088563816,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"9516c5a93553866548be771ff3f72ff1431c03629c153f125b1ecb3d450cab4a","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"31ac0d967c42a3299cd27ddbb6d62c03b311719392f49f15559f61605066886b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"f993709861eedced1591347fda116a344004bc742442bad9d8004837fee55ef2","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"008d7f8d0be687b1f2e521d03b09438833f6c60ca2ce6a0d18ff736f21c1eddb","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"c7cb284e87d6d428e613a27012a2bce4e68717f75bdeae9cd102000ca6398a60","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"23140e5ad24d82a8f70edddd3be915ed999a80c12b783d4b94a78199844208dc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"d3d114c22deac0b3ad76ab5bad960b5f099b5af076b076b1384620f07ecf6355","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"655ecc9a620343763eaf1b5bd992e7a2e4e3f24c4d1f7034ce66fd4624207a89","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"1ab064c0339a3f0cd00e41fce22bf8eab05175589a4246995c08219f9f1a04d4","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"5eebdf409ab29daa4047fa989038a333fa1c625cef2adcb6a5b1f27972199018","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"772d60e03a9d05ce746689a67375833eb3bc49cc41719abd91dc17f01b1619ea","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"0abf0c64a9ead3fe4209902af9f489d83aaaa23c9e8e650bda1bc47250498fc2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"497c032895215ef8e7058610d20d0e132f267db9bc1661b7700f55c568dd8f23","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab8a90d71a2768c90b5158fa57f2395d4996031da73c96d28166b915f6c0f709","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"b1ea619032a41b4c9ef76c609451a8083e102c113b0acbb51dfde13e3e9e3448","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"f5626ace76e61bffc5312a61dd76c9f51f4d712a8196cf69b2665308c107745d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"c12577380db587b1b0b2aa784a1388684a8878fd82aa81586def46c91bfb5013","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"0ffae107db622234864447ed413809e48a65930884838b7e37340b1419874155","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"01a01346c2566b7c673847ad1cd41355f2f3cddefad072fe8e42ea5c848605db","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"c565c95da4865cd9f305e6ae81f6a8e0f7528b4e0763a20acc786bc5726cb459","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"078f9d3b516be081670b90c48ea011529f3a35b38e4ec66f81a2268c439d6b43","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"ad33600e34b7b12cab3930bf41e57fea026efc4555182eba2bd6f7848934db6e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"2ac48ab25470b3e004cf547d46bd5c1f7c095e5b81bc51d3d4eddb4137c39f4f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"e0af8f85baf8c7e3e46acbfdc8acc140de0ce5e747d0a8bc4a322a96ae5d29bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"3b6c99cd04bf37f0306c6f2a3c981d5e110a5a71fc7bc82dc42b65d882e59b1a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"dcb042a3cf3ec740417628cbf07d9e9624c206a40949db0dbf8c412af2b94bad","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"3a8f45cb99aeaab604ebd57485c796c5377c37e95d583970f1271f11aff94257","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"acaf214d7ea672ad50a444233ee2268627bbfc6c86a4b60de04e1489b7e896a6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"3363f88541a3b1f19d0622a1dc7124ea82f851ea32b3b857d8d61894bcf3f16f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"be6eb29c7210f77b5b4accacf1a77585d88e2bc5a5a8c784e8d5881f8072de7b","bytes":5310}},"requestSha256":"6c3b47d95e0a9b20a61ededd0bd0e260e2bba4c544e485b636628f99b26fc377","metricsSha256":"8b8e9d869ba6c71ca9dd1ff588d136e5d858efa5e39c224d6dac1f9a9b8fe05a","continuationSha256":"30590968d25afdb8df2037fd30f18ca11be16668f95512a425b41b0f11db38b1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"078f9d3b516be081670b90c48ea011529f3a35b38e4ec66f81a2268c439d6b43","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"ad33600e34b7b12cab3930bf41e57fea026efc4555182eba2bd6f7848934db6e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"2ac48ab25470b3e004cf547d46bd5c1f7c095e5b81bc51d3d4eddb4137c39f4f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"e0af8f85baf8c7e3e46acbfdc8acc140de0ce5e747d0a8bc4a322a96ae5d29bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"3b6c99cd04bf37f0306c6f2a3c981d5e110a5a71fc7bc82dc42b65d882e59b1a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"dcb042a3cf3ec740417628cbf07d9e9624c206a40949db0dbf8c412af2b94bad","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"3a8f45cb99aeaab604ebd57485c796c5377c37e95d583970f1271f11aff94257","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"acaf214d7ea672ad50a444233ee2268627bbfc6c86a4b60de04e1489b7e896a6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"3363f88541a3b1f19d0622a1dc7124ea82f851ea32b3b857d8d61894bcf3f16f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"24a86a69d84c33dfdf4f8199ce0c386720569559bee846869114f4bb922bbe02","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"13ae69530f08a05e0c545074d75d9660a704d0027a3806b2b14854190b28727a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8dbec3e35359f2783ab9822db33ab167effb71c8414780394a77534afe98605b","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2480673","recipe_identity":"6fe24bf5d539f7a65af01f604ecd2c1be81af9138f22af7d41f756c284425315","loss":3.5912551195912434,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"73ca1f11a8a7ef5cd099f712802dac9ada8628d6524eb7b87dac04202602aa83","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"4400f9b7e9d22cba0deb5165239fa1c5251cbc6df1a6dc2e43bc39940177b815","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"2961ae53224a3ff95894932b5ad1830fda1d4bf598bd96ef810a9d271250a955","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"4ebb4fa88309a1d2f70151fcaf4bf39a7c317d600cd0821278cdbe9347732dc3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"5a48ded14c1d0dc0487cd4256c3de25eec23e15a1ca7edbac601c661f443e205","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"1f65f1c92128a464cff2a10e87eeddd6865aa55ed83abac5c015e81658d2a9be","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"736eb40ccabbe7e6b116e631ec5d74214426f01d74e63fe6a7725dfeb0c0d66f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"6acd18726564b0dcc3d8235bd8a3860866c9cf940688006ed674ca1e19dbaeed","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"dc207d2f793be7ddec1be5a38758796ea7e3d0fa800f23ec69dd9f2d17a8ab2e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"7e65046a8331bb5d9d1ac7b42ab8bfcfcad80b6ead3bd054b81b6c3a4ee42c49","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"0e7ac20691371f8c7c0a31f0a67c46e17e40f2a20f1f8c30e22a834bef97795a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"dda5c026e56b7eed1576ab682ccc3ba1c3bd66f4a0c4a3de049ecbb0ad4f5d01","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"3a2cebab9264ecaf8838e39d49f9599bfc7f44c9d6ac0b163bab6393de1a3358","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"f6b433b5289e50b736dbcd3356376fef4204254f8aa1ad323a3bd1f051feeb9d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"63a0654408e01c8c50ef5831bb916b56518f58db946bc79fc70f725040a5bd11","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"45029041186ea78dbcc9636b97884bea84c7a063f6a471225e1ad23a88253248","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"8db6a4bb491ffd94ff609946ffea156dc2accf747e3f2852dcafddd7a15c0bd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"cd65ad4ba2e6c59cebb2d21ca11f2dac5dbde956a4a5f3051336083812d8ffcd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"4006b74d4b0d1f522f9bbc08db63a5dda66a2c017815dccb080402f124ab2f69","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"906893cc76251d75f61308591e4b179ad22467a6c2d8755bcb3538b9736217b0","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"19c50b240845c74789fe8ebe6d136c17d106c2fb5178caae1c79a96aad4031f2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"5f602604315980e38cc65fe66bcaba4913bd88e31c03d9800703b9d2bd713bbd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"3a53da8d08ff647685e06acb5d32f4aec84f17aed02495ed6858a067b6543e9a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"9f281473c1109489c108357ff365da785b738453145ba4e4ba617b090fac35a8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"c92cfa1b20e0956e05ef195e98c2cc464afb0699cfecf3f80e175e5054ce4f96","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"356841ee229195a4162ce724c6dce40cb36cb4120d66b51ca1cf135ae16a431c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"8c03289c79fdf16dd833321e347991aae7e9510c44940488523601b515d9f373","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"39c16c29c2a3a22c3949aef11d3872798648799036c1757d62bcbdf005a38bbf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"ad2388e57666bb64ad059bdd26d745c9fede11d26d99182e90dac1b642778413","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"6d5313e07ef9d7caadd6767e7686c324f08fa1e969c4c349aa7f21bb9f6edffb","bytes":5339}},"requestSha256":"3ef2a3338fad259c834deefed1040741728af3a6032516cb8564cb6a44765084","metricsSha256":"b31f87fe2ef13d94974319a551a71555a45481ecf092d9910ab59ac05351bca4","continuationSha256":"5574ca82412fa4b8d9aec2917c7aee0c3fbe3160674b4b533a44d050852548bd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"19c50b240845c74789fe8ebe6d136c17d106c2fb5178caae1c79a96aad4031f2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"5f602604315980e38cc65fe66bcaba4913bd88e31c03d9800703b9d2bd713bbd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"3a53da8d08ff647685e06acb5d32f4aec84f17aed02495ed6858a067b6543e9a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"9f281473c1109489c108357ff365da785b738453145ba4e4ba617b090fac35a8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"c92cfa1b20e0956e05ef195e98c2cc464afb0699cfecf3f80e175e5054ce4f96","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"356841ee229195a4162ce724c6dce40cb36cb4120d66b51ca1cf135ae16a431c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"8c03289c79fdf16dd833321e347991aae7e9510c44940488523601b515d9f373","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"39c16c29c2a3a22c3949aef11d3872798648799036c1757d62bcbdf005a38bbf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"ad2388e57666bb64ad059bdd26d745c9fede11d26d99182e90dac1b642778413","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"f516d4ad08a4102970238ed3578c56e28ab083ad1f0e92e9ea3d910356d3a648","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dd8f643ce2d9f90cbbf8fa069747b24192f156edd67b6c1fd7ae65a478f2fc7c","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6fe24bf5d539f7a65af01f604ecd2c1be81af9138f22af7d41f756c284425315","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s42","horizon":40,"attempt":1,"job_id":"2481085","recipe_identity":"021c54f018081dbbda4334d104054ea341460e292ad87a9e173395b6b3b382ba","loss":3.508575881067268,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"6100a7386bb01d2e763cdc1a6409452397c2e4102a47932b82aa43bb2fbbc6ee","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"1b38f7dd690f4adcc17f641bc9afc0b2f7040f8562f716b340a3b29729754c59","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"75075b385bb784cf5f3d2069f536297c3c6db118da9647557e6f8feaa75befc9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"687e6bc0ae3fb18683fa4a70c5fe807ec8e084dc5d83b6fbd598fbb2ce661802","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"80d79eb579552a91ced6e06789d84cc9298efc987b997215a7214e2991ce6bf3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"2fac62950c788f6fc68ea8a28b106af82bd9ff6fea832e46c7ed3b8e3c933ce1","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"ffa0989c3a2ec247eaacac1723978f202a562074eb9e8c68b1e177653564175b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"1fd2db4d69e95f7322190dfbe67fe60bec57924f0f966aad1cbbd146ab9a4617","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"db19b119fae733b20ed7019bc3d6c718f406e6fb465b1cb1677dea6f4f5249f2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"722c26eea4827cf11ac5d0d31c4030b2100d5c9c37e2495dee978a73b2030bce","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"f439413870e509f13266f737c4e8105cdbab5d89935b5714d7baa626fe3ab717","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"dd11eeb6d02973b526f816cdc32c79ea7be447649a10332bbf839b00b0482f27","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"184de996f2a97a5d88b698e4bad5af65669d04769e121094dc9b03f5e9ea679e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"31841fcb8a7c768b086a06075f574cb7d277a99e5da6b72e642b0646d9656145","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"bf63aedd5eba627a86dc4fbf0c1fb6b4b2cadb87beaa9ce22c709b31e6c28d0e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"675fa4d9b70a7a31536209a8ec7883854dcc3f692b0cd0ed5b368e956bea1f62","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"ed863f2157148f5c110ff4ebd53916ed30bacc26cf9753ac18d82034c34d5ebc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"e96eae7e68f9ab7897e4b938eccac34554774dd7f32dcd9f5804b2827d8a8d65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"ffe98a0f9b0038f67576cc3eaad49802349014aeb0f109352770efa4b9c84887","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"28b15a2bb7d11780e4f9718c3ee998b30e7ec219724f497927e2fd3e7e3add6f","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"b670d47f40ded659e8514c6cdd9dd45d8ce74919a507710810414dee826cc2ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"f725c996d5b1a4a3d1513e34d1181ac1d6926c5e0c972a3431b91c788e0626bc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"127e3a37f8a2b79ae646251b66556820d4d9f13ea6288bfae8706acbeab9be2e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"b46f1ac59e0be2e0ffb3b4ff6c2acf566d7ff8182566a5bff5f7454dc6ce7078","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"5002f90001ff4265f96c9f3c4de82cf21eac9ae165687366a68a5d7872a85050","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"734e1f0c8de703414c415f348e12a570068338473043d032961c6ef7152cf422","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"70d01bf9df2e35b008795a7c1ba80d2db5f7edd547b28f6f4bd23fbed22cbde8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"eb1bb4c12ec92d7514762de1bdcb059e0e766f4c371650b7efb2db7b73f84060","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"e0906b1355c37b91f7efa38e095bf45ce99f38182966de1b71bf02301a9c88ff","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"18c5f99da0199229113e78d7afa46a53003cf060f2a4859d58b8d58fbfa5e2c2","bytes":5341}},"requestSha256":"0a588a7ce0f6f3a355af19bb016ac81c1e9a99658222d9fa6c100b0b4d17942e","metricsSha256":"4b36aa7d3bfe78ba9b6289042f165d5ee679aa53ae44ffb2dea81cf39fd9dcde","continuationSha256":"d364ab05dab404efefdf8961d57202ce2e4bc5a82d84385a1a9e73f8fd526852","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"b670d47f40ded659e8514c6cdd9dd45d8ce74919a507710810414dee826cc2ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"f725c996d5b1a4a3d1513e34d1181ac1d6926c5e0c972a3431b91c788e0626bc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"127e3a37f8a2b79ae646251b66556820d4d9f13ea6288bfae8706acbeab9be2e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"b46f1ac59e0be2e0ffb3b4ff6c2acf566d7ff8182566a5bff5f7454dc6ce7078","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"5002f90001ff4265f96c9f3c4de82cf21eac9ae165687366a68a5d7872a85050","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"734e1f0c8de703414c415f348e12a570068338473043d032961c6ef7152cf422","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"70d01bf9df2e35b008795a7c1ba80d2db5f7edd547b28f6f4bd23fbed22cbde8","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"eb1bb4c12ec92d7514762de1bdcb059e0e766f4c371650b7efb2db7b73f84060","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"e0906b1355c37b91f7efa38e095bf45ce99f38182966de1b71bf02301a9c88ff","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","sha256":"900a1b5f1d705c01cf4a21ce64a0503bff1497ceb10e1cc5e8916440ca073249","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bdaeb31ffed42734d1a9e48c48f46aac1d6b892f6893e8337a9d7f49a2b839c5","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"021c54f018081dbbda4334d104054ea341460e292ad87a9e173395b6b3b382ba","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s43","horizon":40,"attempt":1,"job_id":"2481086","recipe_identity":"2f282ec91df84c979b87c0a2dbfadf9fcb4e72c325287d21a9685a34361d9fc6","loss":3.5078378357030546,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"fd99a4b8aabf0b5fc010c16886b8922ac37dac522764b75c3ac595fc4ee8951b","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"ef3f877c9c0220940d6e762c322c01de31e5d271fdf3af633eeada92d35af1e1","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"4a3005e18425d847ad5479d4d2a2f1293af7cf41abe0eccc7fe965fb4b040805","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"64a39c50bab8ac180030fa39589cc9311e96a287b28e814d54939d96f22e152c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"6f4f7db46bd6dd7eddccb8f1139b5d0cc84b111f8675c365459b1787bda250eb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"d51c0888e7ad0d3377d0d26eb051dddee45f3298202d8fb29656a228bb40d23e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"48139fbb08bb9519aaaa8a2c64b85f05af472bbbb0343b1bf34de470ecde2856","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"0c9812a320ac8eedff1a0180df520b1221358b885c60b0e0994d961cbac5d994","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"18d327c821a0edf7740d3c3c52c62887d59c6c201c6a42a0b33ddba265032d0c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"16e6a28fc482c6631e69eaf26957fb5d079cf5779d008b6c576af7b2626630fb","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"a26c71c0b544ad6cfb84c3da259376ce1850f9d8a7acc9a62fbe834ff8c77cc4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"99ae3791b31dd8c3e4f9599f773f2243306ed7abbf12bf062cfe54ff88455e25","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"77088fa45d8fdc6bcece6ef662187e7ccfbd248dfd1bac8e09959d6f4ab62e9a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab7525e0afbd268e0f35556ed23feaa5de8ca582eca95f9af67d9237e7207d87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"71933303b01fb743b36bf4c7e39ff30e525c51167ec0d852a5ea7786a42f6f7b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"fb15417f8051810efdef4973448693e490983ea615f5ce6d89dfe20dac2e622e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"cf67881a46d017385bd1ae3763ffa76875e94cb6f8472d83e6c28e3b800be2e6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"dfcbe83e6ccb2258b61afe6912b6933cebd19e71ec61608cb1408ade80941606","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"f5a84d9e6144b9b39e6a7541a00e9c77c43603523709bc48c9035b31d3e9669b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"5d03d9caa8adc0d68f252ca325e8acbb4b3fc27fea4f3455c89f848d35293378","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"923b573808ed7046d068aa4caab530a26fcef6dfa047eec6e306523c8c11e280","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"6e8ac07e4946d2859825479876a321f5a529eabde5b1bc872bc5953b1bf51c7f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"7ed3d7a7bcbdcb78bce89bd772b600e7c8b7f476021566bf1b5bcc2eff18e11e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"5ff09b62ccc3ea116c1be90123aa8d99ee1241c105b64fecc482f24216737f80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"2a72d97d450c473539d4a5033b43fc0754494db3292c30502dbeee6057e93304","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"2f5bb962c0d6c0edaf7f3371266aac3537acdd00689b5e776507b6d78748a045","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"623d6b661ef710da2b1e58f87230972d4c485c08947c616dbcc4aff03011e28d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"bd763b6b4cbe935946825e1166da362ed2f16934be4fc7e3220e070fe0c9f182","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"541b3caff8ca603420d810ef4a2fd58d54d62bbaca4e29edc38d031b3d20f1f9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"5c59774a82d16c0e6e8cd8d2e737a21265f02e122dea862fa3041a1f634f82ef","bytes":5308}},"requestSha256":"7148142712338f1c7cdf7e572af831ea73b6716216919902a5b387c39219da1e","metricsSha256":"983d2e72776d15a79afa1a59fee4aaa4aeb04f7b2ca7f122c2cebda506b1ce5d","continuationSha256":"0c305a8721dba172e07cc40614230483dde0da38e440855c19dba7b113d3bfde","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"923b573808ed7046d068aa4caab530a26fcef6dfa047eec6e306523c8c11e280","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"6e8ac07e4946d2859825479876a321f5a529eabde5b1bc872bc5953b1bf51c7f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"7ed3d7a7bcbdcb78bce89bd772b600e7c8b7f476021566bf1b5bcc2eff18e11e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"5ff09b62ccc3ea116c1be90123aa8d99ee1241c105b64fecc482f24216737f80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"2a72d97d450c473539d4a5033b43fc0754494db3292c30502dbeee6057e93304","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"2f5bb962c0d6c0edaf7f3371266aac3537acdd00689b5e776507b6d78748a045","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"623d6b661ef710da2b1e58f87230972d4c485c08947c616dbcc4aff03011e28d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"bd763b6b4cbe935946825e1166da362ed2f16934be4fc7e3220e070fe0c9f182","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"541b3caff8ca603420d810ef4a2fd58d54d62bbaca4e29edc38d031b3d20f1f9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","sha256":"7178a7900c92996d72bfb46d95144fea59eeb8a7615b9e4fed80f0a23bd6d32e","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"7f04c528708390400b71e322377f8d9f217433c9bd045bd7728148241154f57e","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2f282ec91df84c979b87c0a2dbfadf9fcb4e72c325287d21a9685a34361d9fc6","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s44","horizon":40,"attempt":1,"job_id":"2481087","recipe_identity":"3f71596b1cc4d6996d7a4ee25d608392b7f74b62127556f45fd94ccbfeb453aa","loss":3.507598741533378,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"21701753a878164c9a646ced244ef6fef76dea53917bf385aeb18a295edb28b1","bytes":880},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"ef5f26d7748daace69a9c7aaa6e5bc87939439b578e192df8e641edf6aae012d","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"507360fc4b5d2682d7434dda38aad5c8a321367b293f270344262bd0f5e579cc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"82c11684b9c7d970782be7882a69e9e4ef92232cfb2a2776915a227a5f80e5c4","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"e2311363061c6eecab28064100268da10600e972eef7366154f3e137754658fb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"0fb8a32fff9e099c49b5f85fcf058ec67aa6099d6764a4a6659a1deae1fc221a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"8d89b24b48a67ea937d59a006ac12851d8c804eba9bd4fa524af72e49efc8529","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"b8721461a3d0b31c750581281bb608c4a8490d33697a6251be96e0579871b0b9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"2e793db238cab83ad839e00fc3c252a784433479102ab56be2dd98285b21e32c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"8dc273a5997fd70e1ef843a728116837df2fcd7a6f50e901aa571379af07811f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"6b41da28f03f75dcbcf52517e962bfd503ea45b52af6fe77f85bcf2d824e56e3","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"d1294ae44a9861abfc4221a86f258693fb34cce5184f6a46d768f63e2faf1c20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"27ff53907cb7c138a7ecd7abc972f85b9f7d41f189c3378fbd543e6975e7bc09","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"bdcf81e33b2331e5b935413167c5c0bb747867e95d1542a10ba10443dae95ffc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"af89e8fc4e5446e84bc95a943e7446e6dbd368f22ac349fe20ff52fd2c63ba3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"c6247543e999b64af0414a7d02e6f164990e3e696e6703bd23b30ef37fda457a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"41efdc84a50c26a50576cfd045b4e24fb3f36eaf269795823af2cfbe110dac8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"58151732d953d5088d43e31e579f8709788913023374800a451124d34a91fa0c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"b952477afcc011286a04b00243f94afd54a3ad83156bb19ad4b4572357e51bc7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"4804360e7db1cab2979b2af59e8792279b30d5b29ee8e0adaf14219cdc3b69e9","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"fcf28f4e72a1ac61872ac5032dafbaff2ddf1ade36be44e621105686d4042f4e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"3d305466dda96952466bd5ff3d6b3b7dfb2363aafd874bfae8f17f3e1df46f63","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"4043971df8ec44df7732211c1b2bc91805b998d811ecec03e1bce460464b7797","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"0e08b6d57455c72e50b721daa04ceae00456a69e8d9199beddc0b747cbe08a54","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"91f87f62a16d074c2c3f6302aced938df99fd3718792260029c4a07b7d612998","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc9147e22a451402a7b381ae1dbc5d10bd07901a5b36953e1d2a595121c4d641","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"5e16ebaba039c8ded44c259a543a767f652e89e107c1f8e487e0a81bee413a0f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"199bec3b1c16c0e71b658f5cd9c39aadb5411d1f74787bf08165038a94efdb1a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"4b7760e4bff5d42ada8a00058423b06eaf46a10574bb979673a9f38081060ff4","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"a61374a25e66a283206387e5fcf8d32704ac9a87b7f212eaae645e38980bc61b","bytes":5341}},"requestSha256":"035255235b12e1e03aa63bf8f8ab91ca728b717e76a477fcd04bbaa10c8e416d","metricsSha256":"8d1be3a3f0381879f51a442032cdc2015ff0b5ad58df550c876e4bf1f603e2bd","continuationSha256":"47c87a5b5f86d05fd6151115ad26b02a24f8b1c2ebb6994982bfdf96283b0e42","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"fcf28f4e72a1ac61872ac5032dafbaff2ddf1ade36be44e621105686d4042f4e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"3d305466dda96952466bd5ff3d6b3b7dfb2363aafd874bfae8f17f3e1df46f63","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"4043971df8ec44df7732211c1b2bc91805b998d811ecec03e1bce460464b7797","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"0e08b6d57455c72e50b721daa04ceae00456a69e8d9199beddc0b747cbe08a54","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"91f87f62a16d074c2c3f6302aced938df99fd3718792260029c4a07b7d612998","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc9147e22a451402a7b381ae1dbc5d10bd07901a5b36953e1d2a595121c4d641","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"5e16ebaba039c8ded44c259a543a767f652e89e107c1f8e487e0a81bee413a0f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"199bec3b1c16c0e71b658f5cd9c39aadb5411d1f74787bf08165038a94efdb1a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"4b7760e4bff5d42ada8a00058423b06eaf46a10574bb979673a9f38081060ff4","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","sha256":"026746d2515eac4beebd0ac1c04e3ccf3b9e3af275252bb2b736007f7683771c","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5bdbb9c80974b08ba729a57cdff4f03b58be4dcc61e81aa45775d48dc588efd1","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3f71596b1cc4d6996d7a4ee25d608392b7f74b62127556f45fd94ccbfeb453aa","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2481088","recipe_identity":"909400517bf7ddc62282f2ad1d63e5fbf0cc91b2ceaecbee3bd13e1197259bb8","loss":3.5070463037187873,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"879154d333c22c7e91dae06600c49bda8c2bcd2d6ee990e635813c7715f5ad13","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"b8770def04ab01d3caf7c849e26ddd30c8ca7614c27863a11e24106bf0c9b06f","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"8e97c43bcc9aa1c47837bac2309abdbb27f8b59f78853df0616b91d108808c0f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"8af4ea68347a8a9e6c6856d60c0eb79e7b11487b1126d7b4b81cc8901edb5113","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"64e3f0384ad73282b7426d10ae712e0baa1e80202baabb660bfe9b7fb19abdd7","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"b146798409d9d40ac73f803e5a44b00be85eceb3a7253240c4fa535a85861f0f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"9e8f6806997363de08c333941dcdd4f451cd3654327d547a090b98f15ddb34f8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"45be72d1f7ef8d776a753e6d2f627d581899a0b16780415a8b0b076248aedf3e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"e222615f45843613e49371aa3e896933296c99faa22b3366b1f19e04255b3781","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"1c45dfec43eb81fb786e13b69c98572fba221bb8fc06faf154d4edd26d77d02b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"062455105079354dbeaf85d04f8dbd90740965a5d741a3ca3c93bb58ab8d7f73","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"3d13025a0a3908ee3b3f758bcb1453ccb8a8112e025d55c26a75aa901764f3a7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"50cf8160f4765c8250d04eeaea53d9da61298fbdfc5720502328060ae65b3822","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"83241f5276cddf8a145d5994cfc352c4f5fc4749461a657971f281c15a944dd4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"faa4c52f7dd49c5641a1383a6e9e12e86b9555fa2a71898787703d808ef583ac","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"06e48d160bab0eb2f76dac5976a108ff2b66163eb56f7a455fac43dda1aefa3a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"8629f7fccad833a64c73be90c05116cd9d43d18abcffbc2bdc786c4d3786725d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"33d2ce68160563bfc3a0df22d1f4e073a70a0feb960ca89d9d61377a5bdf3b20","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"67fbc4178cfa00e5b2033afaf2f718f5253b4e79a17674ea11b9c31ac369af7e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"061cf029e03bcc4e36aa5e48daf1b570352b99f4f0c54c3ed80e9d965eb5e9a1","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"e7755bbbeb7bf7853a20e381cbc60c69ffb084e5ce50580adeaffb14fecba5fc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"1a9a73429c58d56eb0deee75af7cf7d1af4f37e6492fac2ec09c239e0521d358","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"cfe47560b7a5036b43d073e2568ef04858a8dea95c52d0bcb43a93bcd7f1720b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"27000aeeaa975e130f135ec4afff230e0f9324f7dc411fa9fe18aedf14700c62","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"230b156cfe82e3e3d95c3e31b850d4b537af845501d757fbf7e95d0450174b0f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"00e1e3a835824cb693f871b405d1e17c29dd87116cda2833387dec0b7f98e513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"1f33100c27256cbe32e3c3f4775c87396f29aa4aa4cd39b71026f1fae3ebd9c2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"ea610953962ebfdf9c6be8fc19e0fd8b1e7c6850e6e9778e4d153c34c987d115","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"d8c3440fbb761dbbb6b7a8d96721e3714ed4c35b8750ed4170cc7f7b3ed42eed","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"23f246816640934c38c3e775ee94a0ad4f9ac82422a9765bd75f9c339e7ca19d","bytes":5254}},"requestSha256":"4cae6a15eb107935ac1c34261574e6666e9f0e2e4dd50c02434af8405e48730b","metricsSha256":"a2948acda7576494fb6006421c9c267804935e232b0e5c8e9ef9f41b16bd54f2","continuationSha256":"aed151cfc1b1cd204ae6663823b9780ad598dd3a6f8960a7c96b29ac2fda31c8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"e7755bbbeb7bf7853a20e381cbc60c69ffb084e5ce50580adeaffb14fecba5fc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"1a9a73429c58d56eb0deee75af7cf7d1af4f37e6492fac2ec09c239e0521d358","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"cfe47560b7a5036b43d073e2568ef04858a8dea95c52d0bcb43a93bcd7f1720b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"27000aeeaa975e130f135ec4afff230e0f9324f7dc411fa9fe18aedf14700c62","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"230b156cfe82e3e3d95c3e31b850d4b537af845501d757fbf7e95d0450174b0f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"00e1e3a835824cb693f871b405d1e17c29dd87116cda2833387dec0b7f98e513","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"1f33100c27256cbe32e3c3f4775c87396f29aa4aa4cd39b71026f1fae3ebd9c2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"ea610953962ebfdf9c6be8fc19e0fd8b1e7c6850e6e9778e4d153c34c987d115","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"d8c3440fbb761dbbb6b7a8d96721e3714ed4c35b8750ed4170cc7f7b3ed42eed","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"cb0a15e2ca6a4d6dcc88dddea1bab0684d5dbc8068582474364d9f17e277e345","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f7163331ba61f1e9c7934a822571d34e39a5d017d29de9a7f48ffde9c0883f56","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"909400517bf7ddc62282f2ad1d63e5fbf0cc91b2ceaecbee3bd13e1197259bb8","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2481089","recipe_identity":"082147df37807ff31bca126f8782c53f2024960b7a167f460f75ef9fb323a5fd","loss":3.5105572381927987,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"3cf90a919f55e33d0b3dd0c048404c641e0f90caeb0ff862aa123d71ff0d83aa","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"b7625ec00ca99be039398a9c6a33f88c1cd8c17fa6685db7584448a5682dfdc8","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"74170c536dde890f82195416e302025c7aea25cb5a19dda5da4e2aa006fe6de7","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"c617defe58e579669f8c69683a7346cfbb37010c324c0b2854e398606ff45d9e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"a64f5cd5e6929962f41e74c432afbc4535472c579c57008685b308f4fbb765ee","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"aeb1011f9450bd3443764777a4ddc9ac9e53b814d22b0b9a72d0ea52f9d15ce9","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"d3f89025b0dd993f6f02aa2c99fb9a3bb71b9c8e00feecfc228a028e5d7f2016","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"4ef452ffbd0746acf9b6dff3718c846915a1134517fccc1c86081c6f24fcd673","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"802e44ac44c48c586e5e26421d7e3a97b7d7eecab4901d812b891a299f4ed0e5","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"89701ad1b814328058ea9b2311e58db9c182c4224f6653b3ac4194dfa8117356","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"54628be6f6e90f2fa8fde0256f75300effdede4b1793c02e7ee7f90002ca9867","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"8aa0996a2631f75c78233e141f1f81148a67e52e6b873ea72ada8decefd50660","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"5a7a0dc44ecb4cfa1592536d711be626ba927916abf084f975b0474c9f6fc8e7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"321532ffd17fc8e84c5e4c1d33447f24cf4f76d31ed93881b32c848f20ac2336","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"d65d0103d8731a894893f6b349e278a56fc621ccd2754add4dec3b532961e41c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"e7ce173864f848932cf5586022e45f680bf829ebc7a5a62a883c369a46025032","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"2ac129af3232491fe1c1a996ffd8ad23d7369854195991f2912815a5fab7deec","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"e2ae409fac374cb93175277f72533d5dce58fc432a362d68950ac0d1bf402d9e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"729766b42979aaee2958c52f3cab0fa741341ad95303585cd4893f93c8d14f58","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"cf43d06303a2e12652405d2c3d4b408299ebec57fdebee628d11f720551ccb52","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"98b2854c5cbc83ccf9fe96f3bb14691f1f7a7e9592d0c1958518a99647ddf434","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"73e764f29ae5a903cb8307af2fa6a5e05c4766dc6db97c79aff94926f5eb16cb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"30fbd73361c80c3835a11855b69d0d7bc2cac4bc48c4b0677aac667ebca58c8f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"22593d8e25c6be6987efb62301e2ff5f3258799e00c84ce2603da9abf1d6918a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"abef92d6356f1f96566aa47a9fa60376118bb2981dd764425885d24b96f1f2cd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"28e86636bb177ea934b34751aa2e629b05e42d77bb4a71455a887f468ad9adb0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"965f006c5d0591a08532a1f0d141562c5dbe6c325ffe31d470340dad4b6bfa8e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"e6ff62ad33073d2139217d4d5ccbe26c2fa7f1fe6fd424d03bb161dd6acbfdfb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"4c11970f218cb8f63bfa85b623d530d2f63d73e45506d74a0d15207adaf8f2d4","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"33752a0f527bd91e67a8a71bbb225fa993dd865967a21ecbc8b1bd92e3b0986d","bytes":5253}},"requestSha256":"0d3d4e96e12791289a21986bca11e9c721ff6b05df69dcd3bac165e864f6e4cd","metricsSha256":"5cc5a1ca2b80e66be9f131d284a952f373a321605066c6db44e3e7e7c0c2d647","continuationSha256":"0b888be614b95cf9290965a892ae50c813170467f541ab1436ec874f31cac632","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"98b2854c5cbc83ccf9fe96f3bb14691f1f7a7e9592d0c1958518a99647ddf434","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"73e764f29ae5a903cb8307af2fa6a5e05c4766dc6db97c79aff94926f5eb16cb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"30fbd73361c80c3835a11855b69d0d7bc2cac4bc48c4b0677aac667ebca58c8f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"22593d8e25c6be6987efb62301e2ff5f3258799e00c84ce2603da9abf1d6918a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"abef92d6356f1f96566aa47a9fa60376118bb2981dd764425885d24b96f1f2cd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"28e86636bb177ea934b34751aa2e629b05e42d77bb4a71455a887f468ad9adb0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"965f006c5d0591a08532a1f0d141562c5dbe6c325ffe31d470340dad4b6bfa8e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"e6ff62ad33073d2139217d4d5ccbe26c2fa7f1fe6fd424d03bb161dd6acbfdfb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"4c11970f218cb8f63bfa85b623d530d2f63d73e45506d74a0d15207adaf8f2d4","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"da10691aba2648aca02560350cd4e79762a59cb8f811fd4588096ce910ede2a5","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"69d46d065b831692a48e7a7e44b15086074f96d72203ed94d893bed0a3a67224","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"082147df37807ff31bca126f8782c53f2024960b7a167f460f75ef9fb323a5fd","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2481090","recipe_identity":"92c453d654c0fdbd4946631bfbfc854fa8d0d2fe44309184590ec7d5477fd1e1","loss":3.5119876925955005,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"76411c935e450e32f72dfc552ab7d5004a4e06c15ae7a2d7b3d45986b91bac24","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"a61e0480333ed637d3df48b0608ec5e584c3a6a5c455c7f77a64503abf0a615b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"61202fe21ae3830195f387d401cbc7f7e74c76c9a77862dffe98a064e4a84f35","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"f1c714255713a99829a70542166358a1083cb04e26d544f1d498aa2e1cf531b3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"06a349a174f7161efe77eaafdda8a70088dfab89cdd44c4097371ece1d623d1a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"1d5e6e9ad37881b27e7d4924b065215244ea96313c7b7b6cadb158c630240380","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"425b1531aaa758409861862ea7810cc1031aeebd693b11ac2c5f4413c4db16a5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"5999ceede89d227fd4715bece475a94622c90ff27ad154e8937f549b03ea626e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"d0464ad582ca6c697a43bda58076fc94266001d4c3c22206a1697025aa262e73","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"5be388c93e8fdd5c266323b71fa6509928b813e2e09f8d923736c44491abb45f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"56863ffbfd38e12c16a7b601a1328f4e760be56da5fbabcb2493a8c39eacd9fc","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"35e19b6a0bb6188724831bdfacbe6b0b9b9df21247240e28ad0368884c1dc38f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"eff43d733005d8d70c2ca3ce55b260f5671ab5226b9db481d7449cbd69355b7a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"593e995f6fc3822ea54d86b6122dde81bfb3b11cc950f750804b82509dc2eff7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"4e4868caecdaffc71a216a64c8c1242fbd140bd61d50142a33b78996485187e0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"e3174fda7989b5f9bb06e416812390a26902d3480b778a232e17fe433f23a625","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"734e9687a5f38ac62de7d01821a5fcc26039d698364d2180e4f18d4dddbe3f5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"ef497ee110a07aec0c401f86243c86da46561191a55ab7cd381beca9754cde00","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"dfd41b27f9f66be780b426d4b8810a04184d4eaa74cf4e5c24c7202bb1790448","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"be6aa8d2426fa4c7f4896b429a7a27137d4f45e10efc0fb30faeed7b015bf2c7","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"d171f71a8ec32621f633d46935a922ceb239470d18e058a2461dd18dff06ea99","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"225ba1bbfd0f6cd652bdf6cb97de2fb4a97ae086966d6102b129272e8b6f324e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"225ecb60bc21be28d5c55d7505e77edf3822f70c2b45a9cfe54c457b284ab2e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"7dc6f062da3aebbbb5b978675d0a50a1b27e6933fb565bbfb60f48d1e5975817","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"b6874a5120c1e6a59660b4b626614eef77dcc83ee0ca43aa8698872b6e269875","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"216c7fbfa1cb9856bd48304da72842253edc07ad71d3f404d02184a9100201d8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"102779ec2ec26459f3e21e4a63605174cbf8f3e1781652cee01cb58fb9e4eafa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"6f82357c83258af95145570ccafd37629986029324cf616bbcd013bd9daffb37","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"f5b6f80ad23f6bcc2c0a5a867431f7411ac93e4fdefa50bd5103fd25bf9d24ee","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"475a1682483c67d126d2e481eaddd3a7bd4464c89aca2c020e8ae3c6f9e72675","bytes":5250}},"requestSha256":"ab10248c29fa9ed2858b1e5dec51a2ae506a67e99323c16244339d380e29b28f","metricsSha256":"c3fe7af94cfc6d00b08e717a9bc19c66a468ca2dc9caf536d5e90fc331516e25","continuationSha256":"6181de20b9b2c4e737f47f89a37dcdbff1721a11b507fb82b6ffe6e4eaedd3a0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"d171f71a8ec32621f633d46935a922ceb239470d18e058a2461dd18dff06ea99","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"225ba1bbfd0f6cd652bdf6cb97de2fb4a97ae086966d6102b129272e8b6f324e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"225ecb60bc21be28d5c55d7505e77edf3822f70c2b45a9cfe54c457b284ab2e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"7dc6f062da3aebbbb5b978675d0a50a1b27e6933fb565bbfb60f48d1e5975817","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"b6874a5120c1e6a59660b4b626614eef77dcc83ee0ca43aa8698872b6e269875","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"216c7fbfa1cb9856bd48304da72842253edc07ad71d3f404d02184a9100201d8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"102779ec2ec26459f3e21e4a63605174cbf8f3e1781652cee01cb58fb9e4eafa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"6f82357c83258af95145570ccafd37629986029324cf616bbcd013bd9daffb37","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"f5b6f80ad23f6bcc2c0a5a867431f7411ac93e4fdefa50bd5103fd25bf9d24ee","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"e1972b5dfb04ab622f261b577d14a2a13f5105e7ce926c755e389395d6a8c163","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8f414e39923d9737d9f61c46b9e7c53382a8504fa035cdaa39fdcce5de0ec68d","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"92c453d654c0fdbd4946631bfbfc854fa8d0d2fe44309184590ec7d5477fd1e1","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2481091","recipe_identity":"1742112f5e28acd09d94ba266a91758d5adfa0b4ed52209df512d82292f23ec7","loss":3.5088096238591273,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"6cb937c6af827ff2f918d6c57e914bc223934e6e8310fa133de058647ebf9838","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"380ac281127c6b089c4f8574b5e4305b0fd25e0f4e374feff15418633733f39a","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"9929260d4da3241b5be0e53154067b4d7e021b278d2e1cd183ca82abd8452146","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"32cea1e0dfa626a939276fbba5b77594773a8dad5906035ff57eae0514796605","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"ad0db038b54e38dd81bba43516db8082ff95a4b95fac2630d7890897ca307771","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"9881d378ffc8e51b39e3cf6527e6f0184b5a048652b024d785b0517de6fd19f4","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"60cec52d905e7839316366168499fce2d6bd2016165802716914025d121b31c1","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"bb96206713372d327080ef34b16bebebf3874986681fe237cec5f032e6ae73cc","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"db0b22de1fccbe0c7f88b27fdbc9373272164764b0205a5ccd7a789a7194d6d6","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"be599a228b76766c99a54d7702e39bf8c6950d60af8f8c23ea175daacf0f08dc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"a2241e841afb55afbd96457a3002567dc7650e66e7116b35a2de24e17e0f01ff","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"21a6b4d1a7259878afca38a141acf9572716ed6ab7d59d5a4a24dd2c4409a7b2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"4e4c8506e9eff24e1ce1eafb874ddaebccf0b1eb4cfafac30a08c8ad1ba21f25","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"77bb1ab5769728ab0d00b9a7490c1214f876b76ab374d47e57263177f0da2751","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"f29a3f1201aad817a6a93e63fe06c1998a9460f50cf22cf56ebdb20315b7fbeb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"a10b4983eea35b25371442961fc898caf3208025a175adde61faed467fcbe77c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a719ce6e6fb1ab1aa5ef60cab7ecd0fc3194cd1ad042b5bd9ece2959a5823c2c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"648faba86ddeb87f2fc50e6d82a59e5b7a8e659dfbef5b91a462913f7c67af84","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"dbc275df46d851bb41aa6d42434767baefccd2da2d0cb472f94423d699bbdc02","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"e72c268461bc09dc8b643e2a1edc91307a4e5cedbe9e53ed944a2164fdd6dc83","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"de1f3135e061c79b8c4f9ea97a08c701a0476d5c95e499b082d809ff1bc20bc7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"9f9830c8afc6a54d9beb0e676a9aec782cfff507b43bfaa3a821908552e843e1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"c1ee97dd236d6078fbd795dfbe2e1eabd03f497f280682a3c285b3a797be86ab","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"fa6b8ac52f917e00b9fd471586400eea88a5aa1b8796f5f8bdfceb6968b4ccb2","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"05310749a2307cdca43123d5151340850b3b1a497278db1ac4ff48f217091346","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"7cfd163fd4c883e094e76b2f1ad1a1b29974bab40d05306380e55a9211d8c955","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"3b6405d0b1c9c1c5517860739cdefbf84a3c4f63b4e455a3880ac8df91188fdc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"a3e806436b4b720c4e87d0154cf5764f653d4e1f9c670cc9420b26003e826efe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"b19838a68635c9458566cd56cc99e2b9fda327a33555a2b140a0b88c3ca6b9e3","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"0bf907b503763cf3f23d071c7d16c930c9d4fa2a346deff080e5c6e2eefbae1d","bytes":5335}},"requestSha256":"923ea7a082c5b3506b2162975a59862ae7e48d940b73ec20cec3b48fb7158f97","metricsSha256":"6042fd054710c41f06424bdd0a49c937329dadc246766b304a344ae4a89240b7","continuationSha256":"eac93a9665a027829a7d606e1726228504d23fcd21766b2e169ee308c78be957","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"de1f3135e061c79b8c4f9ea97a08c701a0476d5c95e499b082d809ff1bc20bc7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"9f9830c8afc6a54d9beb0e676a9aec782cfff507b43bfaa3a821908552e843e1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"c1ee97dd236d6078fbd795dfbe2e1eabd03f497f280682a3c285b3a797be86ab","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"fa6b8ac52f917e00b9fd471586400eea88a5aa1b8796f5f8bdfceb6968b4ccb2","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"05310749a2307cdca43123d5151340850b3b1a497278db1ac4ff48f217091346","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"7cfd163fd4c883e094e76b2f1ad1a1b29974bab40d05306380e55a9211d8c955","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"3b6405d0b1c9c1c5517860739cdefbf84a3c4f63b4e455a3880ac8df91188fdc","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"a3e806436b4b720c4e87d0154cf5764f653d4e1f9c670cc9420b26003e826efe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"b19838a68635c9458566cd56cc99e2b9fda327a33555a2b140a0b88c3ca6b9e3","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"cd0b9489010e6b67e32038766ec1c3939438a3fdb5b27756a72ca0f93281d7ec","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e3d02fd6cd61acba47adaf3e0c32f4f793383ec17128d711763b64a1ee20f8de","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1742112f5e28acd09d94ba266a91758d5adfa0b4ed52209df512d82292f23ec7","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2481092","recipe_identity":"eb8853a22264bd174573eee50bf8d7aa9d960fc095673061c41ab126f8f2a331","loss":3.518283526200785,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"eeea8b54db78da29fc80d4315b056efb401c799b9f4cc490ac834bebdb03b7f2","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"b5f3510ef3b3b87fd0828c32030650fda77ed90faa94af0193e4b450b37cce81","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"78bde8db5feaee6438a3fe1f5c2345d4747d29d8c289c7d2302cba1e91f63c3b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"b1e84f2aad2b8468778293f0892768ecb2ef4ed2986ebb677c56d1ce54b9e32c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"9d9790dff84b28bc1462cbcf3408957ebab677df34f4b214884a7dd0b877b922","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"38839899777c98689a10a87376c4c0e5958961844fe1ad34fd226d72716b16b9","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"3fd800bba97e07e09dbbc4471180fd862bea782791be0540dd2822d732a09812","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"5acce41d6f01e2d56fc6b0bf07584d6e4327f979da29bc615a9c3f391b276b72","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"ab858b1ef8fda36e84ffe7b22d94a88c1ba9d2b60b83f8f628b2dc6166735728","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"043783c1539b78fc2a925ac4839c5850c6e0d91ca854b19470299830a1d7abef","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"0c1b590ad26e2fea0810d0e05bd0e14495e069caef2d1b33449d6fcc76f1cee1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"aa7b6ba0743d89de2fee244348f67740028ff89d139a33b52b2a2f9eecf79ec4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"ad2305cdbad9f1e38157f618d770f1feb8f69b06c48509f8ec2c89c3d89ebedb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"de56db39ab066f31d02f03051e4d2e886e81ecea65b2f3282a8fad1c0599100d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"5d0871a2199ef53a29d63b16a3536633ad0880405564c2735779bf433fa391ba","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"b4f0cc5fa958ce504d8f04e05dc270301e934f546ed8c8c009a1b518b4185a83","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"623a8ea4f5852a4dfd4b5be868be69c4e5019871ac727f8728d2e028e63bfba7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"fbbdb9dd0a3329477adf020322454b4836d118f21645039f68b686a3f9f762f2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"5c7ab15639116cb50f8b76258128f9e72d819e701c7f7f04c6a38a1620f6616b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"1a008a6deeff4938ec1fe46c6a7c9210cf17a10234fb3263d5e3a8425bcfc74d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"f96a925e396d52f48eec76aa9243262dc3bd797b3b5521b7be1e913dd0382bfe","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"683677a53d443729c216c6e38e005ea15ac47f1438ffcb3ddedfd6e1f818ac1e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"e2a4f6101492a09293d40f7795f4b43280398fcd4470a1940dfecd65a4456fcb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"c555a38eb813470486887c20b54e721398925611de6f259ac8e43b4b87cb5b09","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"fa581e8772e628f93ff158f8381749e245a5a01c5b6d4649bd768d4d33be50c0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"701399fe360c45a403b114914d5a856c7f2e6645eb1e6a7c8306a58d529aefcf","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"f503ef79f0880b2e283857a5fd2a87176bfa6c160c5d4d0e178cdb7a34c1f905","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"918cd115b511922cc333a2ce8cc68aaf74a0021f578f1dd7dc0833f404c054ca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"d5c2f28a77655beca057bc498d60e0eacb7c73ff9e8aeefcd685d9226cbb066d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"fb6a93f235ea5278344d5bcb74c6086b6be368b39940daefe463649294323fc0","bytes":5334}},"requestSha256":"4c7863edf7ab741e4548775e0d641b613f89deda2c5b1bbc575ccbf0d5482a6e","metricsSha256":"2ea4e76daa441bd4a36f24e0219e446253df64366d3e581a32422c0da4605d3a","continuationSha256":"6ece59474566a10a76ad0adf4930ff5bf2e353bc14788c43faf2865ec340427e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"f96a925e396d52f48eec76aa9243262dc3bd797b3b5521b7be1e913dd0382bfe","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"683677a53d443729c216c6e38e005ea15ac47f1438ffcb3ddedfd6e1f818ac1e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"e2a4f6101492a09293d40f7795f4b43280398fcd4470a1940dfecd65a4456fcb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"c555a38eb813470486887c20b54e721398925611de6f259ac8e43b4b87cb5b09","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"fa581e8772e628f93ff158f8381749e245a5a01c5b6d4649bd768d4d33be50c0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"701399fe360c45a403b114914d5a856c7f2e6645eb1e6a7c8306a58d529aefcf","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"f503ef79f0880b2e283857a5fd2a87176bfa6c160c5d4d0e178cdb7a34c1f905","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"918cd115b511922cc333a2ce8cc68aaf74a0021f578f1dd7dc0833f404c054ca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"d5c2f28a77655beca057bc498d60e0eacb7c73ff9e8aeefcd685d9226cbb066d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"fddee2770bafe6a6a71fbf4b95d497f15cd9385c80ae92c675dee5164c4e9408","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d3c1dbb8d488baea40bc613953ce010f2239aee209b001773709f1180442a548","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"eb8853a22264bd174573eee50bf8d7aa9d960fc095673061c41ab126f8f2a331","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2481093","recipe_identity":"f485ff21655ebc865c517f4b7ebe593a9044daac83058cb3b7c6c027c365140e","loss":3.513770410851948,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"c9b24bc8f683e7316102d9582da7a61040e9b783e19a6cdb8afcdd817460c4eb","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"1ffde70988b18966531fdac8c995cde7bf864c8206c2fe70e53b8365cc0963d7","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"aec16430c74dd71704e3103153ab947d716059c47c5f32a70f625df88fb15868","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"891cd51ad45cfd26c82b22f5cc516abb455b2b23ebf59df873cdf10f0c991e8b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"71d25ecacb7d3f73bf64398f9fbf2fab07c56e9b7f77baf458e38353cd6ca336","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"af914af2b8755fe3d1ed892bd04dd2aeecf100ec9bea01a6e253f67fafbb7529","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"ae0df6880122f8aa8d20cd9d200344fb93f0c31bd3607b71bd6eba7e8d861ba1","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"ceb3bb95398b139b46ebbe383716269984029a9abde0dd02d676b2b8eff61dea","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"5a74f662e7014d7bf4b8cc63a971bffb5c2e02e05f7f7a7b96b4308892863b83","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"d0221505e2e606c7a274c13952a144bd530e655ec6c572d0d8693aeb9670997d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"cc8a58fd5d0557af385b9a9bc91ca344b122561043d23a0335791e7ebffed873","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e9fbfdda445c85bc29a7e07cf54f9d670b372862299df304b0c594ed6ad7940f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"f79f144c58fe504441d2922b73de78e1ca93f167f7c22584ae526e5cc9195caa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"090e8b11a6450a30567413a77e025efc409f35d42416fa4a27d3540fbbc32710","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"dfe0923f048be2097d89683fae65dd5e84bd9844a6e1aec51a4ae68520e7f1af","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"99ed1b3d33182af3bf2d8425f501745467dd902e3f7757d15f06a25b448ada41","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"486c1ef641c62a561d549b6acc101ee61feb772ccab8b0017dba9838ae588165","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"e11921e1761ceb8273ee0aa4190ab30b87cb1228fbd7a534ac0545e6f3611eb9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"e6d564ea40761c4bc927e64dcda1b2285b69351f018a7b13face5cc5b1625bed","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"92990ba3caf606e0a8d9152e871456d7d5be23ba54dbdd5941f23cfaa1bde75c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"36d270ae712b4f2bc4884b719c13e091c7b9989d177445d99cf7f79e061ddf76","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"4ea3c2a9fd883cae5bcffb2d758ba66cb37e7cb863b787d05bc2cb15ebc21425","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"39f83641ab9e8ec5f8ed5d5506bb70bf718e7ee637a8bdb9b26b8bf323834f7d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"351cffdcd7dfc082fb2e153d0c1eef6ae7bd9f5a5136e4cdd0cbbd33167cdd3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"e79a8ce4eb95a4b6af2444330c7c9817a2b97d65905f55538c876f7c3a424f00","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"408b9b059b621c1493989b270170c0f419189b4a428407a3af30fe12f3048996","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"ad891148456a307b360a91f898f18ee0397d13e14e884c3617c83b0f5a48acfb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a95a8a706138cbe200b0be61432ce1fb6d7984057e31ddaffe2fd9f773435ba3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"c408f48a63923d5d1cafc71572fb3fa6e6eaf97ed80d21a265a820ea6c3427a6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"b8fdc2c06c80c4056b770bcafd93fa0d36f9d52fc76874229407136366061407","bytes":5312}},"requestSha256":"329cb0efab2d272b04d2edeb8bf7abeefb599620a57148affbf558d80bcdca23","metricsSha256":"cb6812cc8e3cbacaaf8b6fe57f19b2858f0f1d7a0de3d4275fff5f4f125b7410","continuationSha256":"cfb28a00c943b73031ac99366123595da7914903cc2fd9f82be5ea1e3bf68066","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"36d270ae712b4f2bc4884b719c13e091c7b9989d177445d99cf7f79e061ddf76","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"4ea3c2a9fd883cae5bcffb2d758ba66cb37e7cb863b787d05bc2cb15ebc21425","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"39f83641ab9e8ec5f8ed5d5506bb70bf718e7ee637a8bdb9b26b8bf323834f7d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"351cffdcd7dfc082fb2e153d0c1eef6ae7bd9f5a5136e4cdd0cbbd33167cdd3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"e79a8ce4eb95a4b6af2444330c7c9817a2b97d65905f55538c876f7c3a424f00","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"408b9b059b621c1493989b270170c0f419189b4a428407a3af30fe12f3048996","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"ad891148456a307b360a91f898f18ee0397d13e14e884c3617c83b0f5a48acfb","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a95a8a706138cbe200b0be61432ce1fb6d7984057e31ddaffe2fd9f773435ba3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"c408f48a63923d5d1cafc71572fb3fa6e6eaf97ed80d21a265a820ea6c3427a6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"9c56a1dbf102db3b6bf1ff733c6a5872f685b9d55773f76b4cb44c9a7e41fc72","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"58fc4b5a07c72065c270177ca9a8818e281fc8e54c6ac6e8798579b51cd74dbc","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f485ff21655ebc865c517f4b7ebe593a9044daac83058cb3b7c6c027c365140e","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2481094","recipe_identity":"59c1f97dec45d437e3ced0d08d7e9cf10d557010dd900ec9a422a82df2ee5e7c","loss":3.5451323272193713,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"fdb7d0bce7470eee908f1f33430ceaf3e183547c8275b2d9df5d81b632c72068","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"6034a19aa31b6c7e6c75d6b59479cc348653490f3ebe9e8c4065e3aab83eea78","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"f804e468140185c5e383658dc1c8354b8a2f4def97354d0245c6254754e6b151","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"0ebcc8c915ecf84a5f945cb087448248d17fb6224d82aa116620f33c3893e4dd","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"27f38545f8dfc5d512c8f921786f0a6652268977eb94a64689ab856464443623","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"fbc7e5251d36496a83f4485a88973a77c6750706c112bd307e2a4f8138d64415","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"914bc52bc720c1334b3354f0aa29fa96124fa9078950e6f13194360f938c76bd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"c7a64fdcf5fcb8edc251e335435d6cacfd0e76de686c61389255c70c1b6afb10","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"60bb6aa4188ec1f537ccbf4ad98f553c8731563e1baafaa8961da029de74e929","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"01f67d06db98b7838cc572f07c6ce54628d1f1edef590eecaf982f0ad3624436","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"ccb2d02bb417fc15dbc008dfc4aea17683515b291e6559d7a1fdb4069a805eab","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"e5dc0e0c60fb0fa57067b1c41606f823a689fbff49245c7072722dec96e545b3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"0a5a287258a876ba88b6a1cdbdc6a7e1517ec58899a1496e3ae88251e2b51201","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"7e4718040db83d13a497a08d70b4fe572507401a363d734b8ab3c80fca0e87a3","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"2090583b6a388fdb64f3bb0967ca6d5a3f6c8f0eaaec6bd8220eeb1d47b1c950","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"3f4b453865119a12c11ddd2565d31703f8ac4ae427def0da517ae740dc6a8ac0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"1f4cefe0d2b39eec066f3625ee5a86aab37863a6f52f399ffb28f017a5114905","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"89312a784ce1ed9b7046da1f6c6a5ff604e6e413d755d711b0c8b767ee96f202","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"7f14408f427d3b9449fc0039a722afd083aaa6897720a5d9cd6322038f45b290","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"74adca4253d0f045fbf343d198237537230ebbbfe3bb54d14710953f5b5765e7","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"5a3b342a49884306f6975bdf929cadc7e8088eb289f82e9a6f9b7ef5edbff412","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"87bb9f88a9f997578492c794c8cd01b89462734583b61af9258b4c1583bb536a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"7acc77f8458dd0a0ed27489ac4888f44725b598619fe310b2871ba8326a66a7b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"350cc763fcaffb107c8ff5b0e011e0b1c01ecd060ed704dd73774016b3e5c785","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"5417724227b23e70d6b655e66c6d32a8ad112a437b16a413c255f2d2e324976f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"362701167afbee042473651d9c71ffa445d385c1990fddec9cd42fe751ef94d2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"db1ababf8c2a07cf7fefd2a2d20944f331e73afc52b5d3614eb49aff2397a719","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"ab993bad96554a0f2550f168a46ab90040ff9e48287265f59566b390263094fd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"79272d957aa2b498a5bae257478f324a2695a4b11ff0235f7c7fb6c4e6b84d70","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"49058c54fc83b5fb0b8dc14976019b8da8c8b3b9c3556fdf727bd07954c4dbb7","bytes":5255}},"requestSha256":"6c314b4a1b48ef8e115af634bae37418105e0c4b0b1a7c0f8e40ae14546ec1fe","metricsSha256":"6c88568498d5103d9c3c5f95c4b7dc59626cb1fc708eb6c9ff7898f9a1adc349","continuationSha256":"28ff10a516729bb7ee2d503e1dc8c617d86e8c45e572eed0e37ab80ee595416d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"5a3b342a49884306f6975bdf929cadc7e8088eb289f82e9a6f9b7ef5edbff412","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"87bb9f88a9f997578492c794c8cd01b89462734583b61af9258b4c1583bb536a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"7acc77f8458dd0a0ed27489ac4888f44725b598619fe310b2871ba8326a66a7b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"350cc763fcaffb107c8ff5b0e011e0b1c01ecd060ed704dd73774016b3e5c785","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"5417724227b23e70d6b655e66c6d32a8ad112a437b16a413c255f2d2e324976f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"362701167afbee042473651d9c71ffa445d385c1990fddec9cd42fe751ef94d2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"db1ababf8c2a07cf7fefd2a2d20944f331e73afc52b5d3614eb49aff2397a719","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"ab993bad96554a0f2550f168a46ab90040ff9e48287265f59566b390263094fd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"79272d957aa2b498a5bae257478f324a2695a4b11ff0235f7c7fb6c4e6b84d70","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"e8f2fbd03c4e770030c3074178654c3e6c78911d6108fad99c8eebee9bfb956f","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d28361c031e0ed7003897ec62a41523dc30d8f06ee73fa1cbb831f4c2cb8173a","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"59c1f97dec45d437e3ced0d08d7e9cf10d557010dd900ec9a422a82df2ee5e7c","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2481095","recipe_identity":"dd7156a6509ab9eb2075a92ce9f55e748dae4ff9b97b82887a505386b955e910","loss":3.5433717236466116,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"b8ca51ac32858ae0a2693dd248b208973d152bbde6e7dd1317e6b3f5368f274e","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"2044ab814f09b536a4887317422a9b3c0bbb36b18b56b65be0ac85340059cd5c","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"a87748b4faa91cf407b213a160535c96faacc78e7330c34bd9709b24d8cbc9ce","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"8b0c10c4a3890ae0aa8c30255dddf6dd00eea3c87b3446ae07c6438ee9b96b75","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"ca6d3f56bae70da039a6ac426fae7f1c784fe95ba7bb6abb9147ae6d9b900ea8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"84a25845849f1b72aacc46c884b465b9bf0431a3d00166ff585bb9c8faa4a2a8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"93dbdcca51d81d6720009ea2a392cabf48072af8a07ff1b02cbad566b75f2e6e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"b4214b420bce7ca1c6713d20372dd8820c1386b31a44e008c2b2043de0fca354","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"c353b60ba5c883fae6f249ee53c5fcf2b5b32da4e2bcfea8076a2d482bdf1ceb","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"8f00b8209485fba8202483b8ce57c362b5116be36ee00fa907101ce48788da01","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"a90973a69428d1ef5738d7d66d157fecab44ae3dad7923e038fce7d756302e0c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"4b72c393bdb9f69e7bdce3782ec2e64227faa1b0ee3de0d0d4a4f803f6207b6e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"2e757de9d41a416ae46599d41a01383ebd3a78d232831f71421eb7eb62b480b3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"a12c5ede4bb4339c8fda972c67ae0d0fdf4250ba5182ab1fce57b6cd2d61300d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"ba09000ca0408ededf8fb56a253129b7acd6be1790e127d3eb30bc7bb26aa0bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"1eb018df4adedcf5bfbf7861260903e3134799ca77c8842ac9db0902c8ab77ed","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"3eaa3dfca30e14dd2ee830f0da8f4d50b86cc18a56efcf339066fd6cb41fe1c1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"92e36851088be6c13e948759e3d52ed542199406afd6be7f3dea35327d745fad","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"0d0aae86f9c694b7c25c1d92a235fdecd81289297596cf7afbf162dd573a6685","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"96f94db00f0a52e444a9356e3b62c235ec7d0a9cb88c3803282b322c4378dd4e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"cf76539b2bf8268a55cdd6826607005a2ab37722fe550c85f25e19a9705c0f6f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"f0f3b6b118d68b99088983c25b61d937da0e648caf26673f2e05d1c48b89b1e6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"4270f6224dbcf554bb0855e65c27023139d6868893260c39b17b2250319f16df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"404417e1f21cbeba92a9c80886588793470826230fa6ec7edd6a7100cc9269b5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"887f33aa9f095259cbb3f22db6f0b01f03fe7d6608d48baf79d04ed76e0785ae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"e1324246487e3d9874c620a590731285a9e8240eb1d4075a880edbae379be76b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"399be5551e2754ec7f20b194048299378c61f7c8558361e09772907454c2a59c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"ec9aa15220cc7af44dd44cd4ee5a57d65f0a02bdf3b35d2d251532407c63254e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"8b39fe3ceda0832bf738c69b611acc67a9d4c4a23916951d445ae9b429c52026","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"cf0809371c0a015f0d4c8dcdecd12e377db6bc4ac0b88008d9aa10fcb6e3e9a1","bytes":5253}},"requestSha256":"9be752ccdc74b59da340cbf39ec4cfcf915c1181aa1deacffa19316f9fc8e019","metricsSha256":"412319446e6c8fda3780cfa929e5052b27f476c37f76b649d13a4fc6996aa370","continuationSha256":"cb447b4e92978661e9b7f42a1318807a0eebaa4daf38f950e0c1c2b82c358fef","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"cf76539b2bf8268a55cdd6826607005a2ab37722fe550c85f25e19a9705c0f6f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"f0f3b6b118d68b99088983c25b61d937da0e648caf26673f2e05d1c48b89b1e6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"4270f6224dbcf554bb0855e65c27023139d6868893260c39b17b2250319f16df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"404417e1f21cbeba92a9c80886588793470826230fa6ec7edd6a7100cc9269b5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"887f33aa9f095259cbb3f22db6f0b01f03fe7d6608d48baf79d04ed76e0785ae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"e1324246487e3d9874c620a590731285a9e8240eb1d4075a880edbae379be76b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"399be5551e2754ec7f20b194048299378c61f7c8558361e09772907454c2a59c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"ec9aa15220cc7af44dd44cd4ee5a57d65f0a02bdf3b35d2d251532407c63254e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"8b39fe3ceda0832bf738c69b611acc67a9d4c4a23916951d445ae9b429c52026","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"d549fc8658786c3d7428cdf6989ac1953a7464ecc41a0011a33fe7fcc95c32cf","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ff926f67516efd77af709427c6cfbb873551e0254f0318aead147f87b4604ae4","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dd7156a6509ab9eb2075a92ce9f55e748dae4ff9b97b82887a505386b955e910","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2481096","recipe_identity":"ea414fb12070c56c8608295408e865169a0d8d1a27426f7d95c0a26c3a68c350","loss":3.538147812999349,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"ed3482dc2c1d4771c068ba424f843c516cbe7f45220e66a2b9c0165a50c1acb1","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"640b4f31a56275b05b00bee462b64a871ab4aeb9a66b90f75266e49957d7ef59","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"4bc67d6bfab45f7340af3e786d37f15bc9719f26dd12304078f844e1aa9331d3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"a7f96a269c6da1a90b649a1f98203f194b96efa0f8ab9f22431d1e2bb35f14d0","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"8bccc56eb06698179618fcc5aff675fa8430253a541a1e62fdda80b1e105b8a5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"2cbec1f7cdc1c2d2c212aee3f54bce628924ae93c7b4a59c44c1f44d3537e4ec","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"f2c92271eb2a9f08c566d2c19b271e06f2ffce15cf3bae227c23dd6ef1be1f48","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"37d160481fed1d0187dbabdde6ecc93d1633dc2286f4dee4fb591ea29978441d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"5199bf78e1888986d28b0ef46b603631c48540f36efe9fe23c080d39b0532650","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"ca5f6fc4eaf9ea713dca23899ef14813ca5096a079715c9b3473ede9a009f238","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"7be809c0253d358fb75be056af66e3b6333e9788921e5b6cbd35dea02f75c950","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"827102717adf6c28c522b81749322650d67ec6a69b042bbede76dc6bcf8e733a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"a045304b455f348a68cf677f697abc370a9ab73fd2cf21f434719c9ffbb99f6c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"5d354bc822924b54144036e0dbbde1b7b6002757b77d88317f9c7e9f97bf5b11","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"d213f0655e7c35c35624bd8b87fd087ed83861c7c9e9db360cfd8332c6399428","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"a99bd471846ec06edd5346ab2797a1f2511d008267b50b1ced021fd3e5447ab6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"76cf7d4cfbb38252a3e67948c98045cb4385ebee6862d160bd567009576beccc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"d509c1f3970477ddde1da80e60f0b3466989e39cf523159c3e3107e26bcc4931","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"9803771e681df8b447df584b9f2ebff8d6acd0e7b66968602ec39cec7508cdef","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"55507e4abbc089c202ce26f6e1536967cbc1ac3e6b7d8eca0c07502c0164df1d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"e9c90858d2186e7ef66af46aac2938e75847fd541faeb3d95fa12159179ae77b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"ea533bdc984e62b76ac96e22461e865c0c546b34b941f6d2fa6acef9f3671bf0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"0f11b3e419db26367561d8b10cbb4d5b18f293a315352281bd6a3324b8abe1ea","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"7bc1b425314b5d60b1b9cad6f59789f0cbfe86de1d1aed7faf732aa4539ed412","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"32ffb3f19367d55a0e5f3a23c2e0a1a7657a27a7f80391dedfb2ec67dc0ff9f0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"d31fc2338625ea62d25bfa005e0947c481ff3832cbe44daa21e40c00b4d761c6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"53e0567470ab77964f597b491e952563dfdabcda48bccada86ff9be219d2ca8b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"05f0b754996a4f5e927ca2e6f9d1e4a1830b96fc7dc36b3b51b31bacfe56dd64","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"1fa6a717dc70bb6474d70538905aa94a32fb99d11901997fd073232fa19646da","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"540bff0b1ed83dab95eae59713f4ca4f63dd4e972de93e2b036ca04166896b38","bytes":5339}},"requestSha256":"57856a435ef8a951c2960e02efb4030d3fe96b4b528a8390f1ee04f9e64929c3","metricsSha256":"88747239d81f4e30bee2ca6a4beed02696191e04e49e4ef62618dedb7558e66d","continuationSha256":"bd5052fca0de85a35a0eaea21f2b821d795ad463bf857ad8b8d4ce8abd09b801","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"e9c90858d2186e7ef66af46aac2938e75847fd541faeb3d95fa12159179ae77b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"ea533bdc984e62b76ac96e22461e865c0c546b34b941f6d2fa6acef9f3671bf0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"0f11b3e419db26367561d8b10cbb4d5b18f293a315352281bd6a3324b8abe1ea","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"7bc1b425314b5d60b1b9cad6f59789f0cbfe86de1d1aed7faf732aa4539ed412","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"32ffb3f19367d55a0e5f3a23c2e0a1a7657a27a7f80391dedfb2ec67dc0ff9f0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"d31fc2338625ea62d25bfa005e0947c481ff3832cbe44daa21e40c00b4d761c6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"53e0567470ab77964f597b491e952563dfdabcda48bccada86ff9be219d2ca8b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"05f0b754996a4f5e927ca2e6f9d1e4a1830b96fc7dc36b3b51b31bacfe56dd64","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"1fa6a717dc70bb6474d70538905aa94a32fb99d11901997fd073232fa19646da","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"1eb417230592862cdb53068f2364797ba19005f5a597babc33590594624e0af4","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"399f27c41585d31356643a3ef5f9e5295438ce334a92f0a3c16c6c44a824d908","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ea414fb12070c56c8608295408e865169a0d8d1a27426f7d95c0a26c3a68c350","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s42","horizon":40,"attempt":1,"job_id":"2481097","recipe_identity":"e6f2593db657136a1a8e50acaef613c801fa83625316613394a80c83980e1c4d","loss":3.4735672746780732,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"da964fdbb79bd1c1d5a73fb0247297a62bbf7c7cfb9cbfae1e36fa9d2917bd63","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"75926c694c6088c0de8bfa17a527250e5e2d3004c88fea3ac5ceb2fc3521b082","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"9b6f2f91ef1c895cf761c3652a39376df13b6850b36e703d21cdb05677f73588","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"8ff6900da84cb78a64047970960efb61ec24b4212ee496e9c1212739c5f8c28f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"2d6280fe343ca07c1feb327a1894f427a76cef09ab7d3da4c682b26105a74949","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"07f4a5920b093b34b13a43676513f6577cccfd749c6276daaea94f146850a2e5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"a778456a91bbc1e1738ae74c290ed96e16c1f6f7bda5419bd9d7a62ed51f546a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"e8ed018010704d8c2dcd07743bc5b8dd3e7cd83b9446827bbb860cd6ca8a2b6c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"5bed64771a16a1775cb6f85952abd4bbf4c5b37f685851becef76867802e38e8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"d76633b7bcc7c6fcc0946e634a3348fdad0bb0a543d6ca1c65bd89cf3daef047","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"83310da78329b3983cc5d9c7738964cf15b0b7e6233a608bb36c894b06b2e378","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"0ff4bae7a842f35c0a185fc93c8764aec5f540e0dc1078aafcec28af6b47b07e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"e0dbe30f061030a304d36ee4bb786c434169a66a80cfb2fad9aa1dfbb8cd0b85","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"2f3fc1b9c3114c138a13cd23c3fcc48efef08215877789a1c55dcd9f00f590dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"465f065871b7ff354fe7b2d7dd979600d1d22c90cdc34206edeb57b2ec571a08","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"1743598b435d9565ec003cbc9633f31cdaa99614c57733024a20afdd4fc004ea","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"e2446846adf5b1afae061885779b994ff74f78e7b727767cf0325b963e0c791c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"8cabe4fdf6bcfe03f6ef32b892ff52bee7f0af3759f4fd38f06f3701d17cc93a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"f4317b2d34c247af2681ba44ea1025b54486b18c53e92f1f7333017c06bc3b94","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"c2af7a0fe016db28113ca7f018f0d0068e627ec6a21e049aaed2d86092265296","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"66c532162195e9718a70392b09bf8d1dde4ee42ff114d09b8149f942d0933c2e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"7b123f5217dc537c2e6a290633784a82fa929c7b61a47f9b88b4545eb079974e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"09c632516286a0e6fbc106b7911618f1f203f157a66ef0847ddc415b13ece91c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"dbfa56bd13a3d4f93dcf4e80629424a3544442e4a3a511d45403f0465215dd66","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"fcc618f6e2b0b5e03dc9ad9980fc2de4497116d4e121c4672a2435cdd4d0f300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"435acec51bc99985f089bf0b746b550672f8a4a20997935c1f4340846871936d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"9b5e58662d265f245671e5e7e9a74320b7a65b6a71018180f1a80689f598e2f3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"6fb2a1c7907061aebe9a6d19fa65adcfa4ce1b51376151e57f618f99ebd21383","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"40846ed038126f0293518d4ca276639692e51744b3bd176bed695efa05d923ec","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"56ee5206dfe117ed43b8a97fbb8fd03805cb59c0cf34915c106bc50c66e39675","bytes":5312}},"requestSha256":"c0be07bb3241aec88b59d45e7daa47e6fb5c49b5a671782764504c110bc69e03","metricsSha256":"7d71a81381b025a5880ab234810cede74f6e31dba63e1987300806bdcd8ad9e4","continuationSha256":"2cc54162039e053d4533a415571ac54d84f634f9e1de06c79597ab2e6c39030b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-035.pt","sha256":"66c532162195e9718a70392b09bf8d1dde4ee42ff114d09b8149f942d0933c2e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-035.pt","sha256":"7b123f5217dc537c2e6a290633784a82fa929c7b61a47f9b88b4545eb079974e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-035.pt","sha256":"09c632516286a0e6fbc106b7911618f1f203f157a66ef0847ddc415b13ece91c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-035.pt","sha256":"dbfa56bd13a3d4f93dcf4e80629424a3544442e4a3a511d45403f0465215dd66","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-035.pt","sha256":"fcc618f6e2b0b5e03dc9ad9980fc2de4497116d4e121c4672a2435cdd4d0f300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-035.pt","sha256":"435acec51bc99985f089bf0b746b550672f8a4a20997935c1f4340846871936d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-035.pt","sha256":"9b5e58662d265f245671e5e7e9a74320b7a65b6a71018180f1a80689f598e2f3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-035.pt","sha256":"6fb2a1c7907061aebe9a6d19fa65adcfa4ce1b51376151e57f618f99ebd21383","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-035.pt","sha256":"40846ed038126f0293518d4ca276639692e51744b3bd176bed695efa05d923ec","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","sha256":"929614400ffd8c3199bca5260530ddd6f6d03d4e3879a2c01629ecef02b19ff1","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"96123d836996f5486880cee3654a7e6105fc12ee498db3b6b4cb6198ba32fbb2","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e6f2593db657136a1a8e50acaef613c801fa83625316613394a80c83980e1c4d","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s43","horizon":40,"attempt":1,"job_id":"2481098","recipe_identity":"43dae8aafe05e36d73781226b5d94144d1b899d74dcf461905204cc1f8cf9de6","loss":3.4789400999554836,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"f32f56bf717029e918c9dfd0c6bc6f350af52d71839d8ab981c8aef6bd6f1a84","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"94c351f1cd6044cad2ef38eace6c46fae3e58cf26830fc9d9f97e5f28377e504","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"a0589c974f21de855f0af662fcc3f3468840e6240fe105212814367193b9999b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"c16648c95f001b5fcd26f21364fab1eb6723aba16b573157c7064c69fe68f139","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"37fcfe92c4f8bfcdbdc5726f51632777508cb8603584a3facac97c8550ad98b4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"d3e17c01cdbc736c8d4c8ad792233327db71aa30b78e2e7d0d910ccc5aa2f986","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"8998985073ece26db366cffac4ccd1c155ba186bd9ae020c2aff34e13c11acc0","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"49e84d24ee85d714fae85f8f8ead3ae21238cd2fbacb24a7e4461c300b762c75","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"4b6401a1d8b7f831d3cedec43f05bd651f92afa1a2aeb7ead47a1f043ecdfcc2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"ce34a50058605ca1ada9acff13ec30a648e38773287a5b824515888a95de0b21","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"0e01a0ac6600503d08e4669d53c26d7f2949aa8522e1a462202f7b0c31b8675b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"930f7093adc05f6a935acdd593262e95895cb2190b2262feafb9dec1451fe5a7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"33ecff47fe5e88c510b98e0e084e18eb1f5afd51e34e9cae972b7f8ba2644c9f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"2de344c73bbb43f64ce835b047a024c5fdae19ca7d4e6f1950bd9fd7018a886b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"5ccdf7d97059fc8770cb9254a3b5ef0cdc5433f4329442bd532975fbfc0372c9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"2c2012f6987bd8d59525d00949c7d23f1cc2a842203e6abcd8b3052ccf92ade1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"17aa3b5880035ad3ef0dbe0284d18cdfdd8d7df4a5aaeda2aefaa7cc08258bc7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"16f53742e3265ce08577beee349f621c20bc8008069432ca3be009ff43f375c1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"d8b4c96133c468f07d1a1016b66e8910ab9fe66f105dc8995428a198da8e9af5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"d864b740f9835ce6ca025fe4dfcd9810a15ac277fdd4b5ed3f04f2352c76c03d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"580a3c0cfd5cbbeb136c9cbdc582a4e4d5d172b9ef025b69f3c184f6a267c86e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"ac8e1eee6a6220f1443e0d8ffd4951b44afaea2d2910a48741fbd0b43903a630","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"972037068a1ceb7db79c1c1df82624ca54bc296ff188bb496f55645e214dbb3f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"f7807bab04680a4550e024a5caaba297e1b90b27ec57c563f63c11b0b585f781","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"b0b2e74c35c5052738b376d9b8a1f71ec7ec6f4493996dfded75ada3a0b5000b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"832a6d05a6a9ef0b7567e6e93659f2dde8cd9e1e1e89bfd6686a0c3a26ee1d80","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"b65f72a8841a446f6c2d5d701cc2dbbd75b4cef03b7e10a4d5fda22ae5698c43","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"ac2316e94075e7475890eb0f863e9a8d5c096f6bdec86bc3097c51adbb4ba9f2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"37633ae42a558955bc2086a08b56ae2699b21abad4ed91675a0759f146e37351","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"ab11c56f45ba8e9faeeb5ca5324e095f532004b1e20f11724bf59befcbfe1ee2","bytes":5340}},"requestSha256":"1d0cb2747e1309e1fe95d22bf5973742bc6cd73e8ff4ad84c673f51bb84058c9","metricsSha256":"5974290ac62575c9b65d24797b0c07f66906a15817711be5c8e2499b708784f1","continuationSha256":"166c04d15719540665ba0f63561cdabcc870699e32758ee4519dc78e8e300dc5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-035.pt","sha256":"580a3c0cfd5cbbeb136c9cbdc582a4e4d5d172b9ef025b69f3c184f6a267c86e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-035.pt","sha256":"ac8e1eee6a6220f1443e0d8ffd4951b44afaea2d2910a48741fbd0b43903a630","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-035.pt","sha256":"972037068a1ceb7db79c1c1df82624ca54bc296ff188bb496f55645e214dbb3f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-035.pt","sha256":"f7807bab04680a4550e024a5caaba297e1b90b27ec57c563f63c11b0b585f781","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-035.pt","sha256":"b0b2e74c35c5052738b376d9b8a1f71ec7ec6f4493996dfded75ada3a0b5000b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-035.pt","sha256":"832a6d05a6a9ef0b7567e6e93659f2dde8cd9e1e1e89bfd6686a0c3a26ee1d80","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-035.pt","sha256":"b65f72a8841a446f6c2d5d701cc2dbbd75b4cef03b7e10a4d5fda22ae5698c43","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-035.pt","sha256":"ac2316e94075e7475890eb0f863e9a8d5c096f6bdec86bc3097c51adbb4ba9f2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-035.pt","sha256":"37633ae42a558955bc2086a08b56ae2699b21abad4ed91675a0759f146e37351","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","sha256":"9b9971c0194449b14ee5370579f800500bb9a1b4d9186563cfc68c486329dc89","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"17b9da036f44b92b3f9b774b83f43a84cbd323aedac1ef1f2469fca87fa3cd25","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"43dae8aafe05e36d73781226b5d94144d1b899d74dcf461905204cc1f8cf9de6","loader.blocks":797056}}},"awc8-wsd-h40--0.006--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s44","horizon":40,"attempt":1,"job_id":"2481099","recipe_identity":"3db56bb253af1abee8ac4f594bb4cc7934f1ea50afd9ee571e1a13d8e7f4ba2e","loss":3.4763383999425486,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"fe169dcfcd39a4c1831edd7419b0f79c4bda4e364f49a3a21421d9375c8794c5","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"3376f5ccab58fbfa27d62691f37fdef5a2daf8ed9f671e3332eb737fc6f44588","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"c9d469a79126e29ec7f6b8e059a97ec47ae24ce79868a88c787c00538264be74","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"8cebc5152f6e03aa37220b7e73ba8cdaf04de892d7530d6a08ce68dcfd2ef704","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"e8e2704e15bec85538ff21cd6b64d2aefca6e6c76e6e964bcb3ce0048dc08806","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"b1f7347def5e8db4c2b34832c34b5bd5ff5fbf3d204f8ecba3f1ad7eef946183","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"f3188d98df0c7623960992a1491cebe399ff99396a5d9b6159d86b0d7b8a7d2f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"6599b8b97052af5a7ca23e8b80e689f0e76e4451acb87836724d9a9bf86e0181","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"4b56c3e4e9bf033a2e128736b5b8974a99ae25e3df5f8b9711ff328909748ebc","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"1a7ad18867e0215e1bcb92d25c9ebc3d7563f2fe6d9864cb64343fcd0f76b773","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"7f0593104aa69b2c954341cb122bfcecda83a4966bb0f9b1e9c1db683782eeed","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"724fdb8dd67c02cf1abda03f4dc09418ed5801f1b1c5fb2869e0d39e1887a795","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"41c44d0df20eea437e851fe67d949b84b6c6888898b79236a835bf3ed6d92d3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"f9a08c4cc932f3bb30a9266955c12cef9f8cb35ec761332a4ecc558426a55167","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"fa4c467be4aef10ae48954db67a362acc5df52bbda2dad5829309699c89dcd36","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"e05238fe1c6f6adf84e8fa5a52f7b3106ee2746d5b880f408557ebf5f6b68888","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"c4b0616483ac2a6f7a574f2145f5f8b6f95b53b1cd935b20a3478e5c91f07d67","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"087333f504f89424fba08f214d0fc36e26c9d55111d1403c158c6fa69df61b2f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"f32816ad635cf020be3a3a2767c56b5996caa3db6f4ab8b326863c97900438e9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"2c4061505f25db7f891a2edf004dc63e3ce4fc993e808f23dbd90ff699fea811","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"4bdc97f8d4e80b5d4e6e5a954cbed355e145b1adacf7c4b8823b77bcd40aa6b1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"6cf894f9a812e34dfd74c56910b8bfae66007ba29e779469c661197e4f82133a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"7affed6c7a9cb0792f0c729b826a1d6d865f89e428baafdb66bc66156c913d4d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"5db51acbc00839c2042dadd31eb3245232c052383f8a365eadcbc68304fe07c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"c08027148e461a1828536561324edc0b55b42ebf6171981eac47887f697a0c17","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"cda178d4824acc44a51599077ed20ed3ca47bf935d88afb85a05be5654925e2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"04d3fd6e887404e5225ab74afc30e5710b98a634cb471d8118805ad2f3b5d453","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"5956cf06a82e1b3454f606bfeb51d21d55b0e6eb0570b44e694b1fc92a07a771","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"310da54d816e25cb710a4379395e2b1d3b5d4ddff297e3f1cddc305856d578f0","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"a1edecd1a1cc69ee5ba8a03f3dcdcc7857b5228f3a34ca71b29579c765a98b19","bytes":5312}},"requestSha256":"192cb65f30d28be139c485807e958793fb50d965fa575eca6908b7a6f7cd4705","metricsSha256":"1bbc74fff9c702404575c242b0aa0c7505fc6037e20d0aa6b0a9900d8408ecea","continuationSha256":"5412d18fb293e3dfc0b27cb06b0ad51afd0391288734d68534fb5492711631fe","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-035.pt","sha256":"4bdc97f8d4e80b5d4e6e5a954cbed355e145b1adacf7c4b8823b77bcd40aa6b1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-035.pt","sha256":"6cf894f9a812e34dfd74c56910b8bfae66007ba29e779469c661197e4f82133a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-035.pt","sha256":"7affed6c7a9cb0792f0c729b826a1d6d865f89e428baafdb66bc66156c913d4d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-035.pt","sha256":"5db51acbc00839c2042dadd31eb3245232c052383f8a365eadcbc68304fe07c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-035.pt","sha256":"c08027148e461a1828536561324edc0b55b42ebf6171981eac47887f697a0c17","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-035.pt","sha256":"cda178d4824acc44a51599077ed20ed3ca47bf935d88afb85a05be5654925e2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-035.pt","sha256":"04d3fd6e887404e5225ab74afc30e5710b98a634cb471d8118805ad2f3b5d453","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-035.pt","sha256":"5956cf06a82e1b3454f606bfeb51d21d55b0e6eb0570b44e694b1fc92a07a771","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-035.pt","sha256":"310da54d816e25cb710a4379395e2b1d3b5d4ddff297e3f1cddc305856d578f0","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","sha256":"1b88354185291276d595c7c20207c5bcf231f7ec147e0e0aa8c011081b3a3ee0","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"80800b13f26159de620f6538a469f84add13c4ed58f965ea6e3fdc61018d57b8","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3db56bb253af1abee8ac4f594bb4cc7934f1ea50afd9ee571e1a13d8e7f4ba2e","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s42","horizon":40,"attempt":1,"job_id":"2481100","recipe_identity":"d1835c98fe6ab1ae8a9112c94dffe867f135802ff1c1bb5f76d1d2a9820d4da0","loss":3.4738049895081455,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"e6c1b52b6aca80e3be0dd7a755fd03fdf545ca64f6fa2a3f1a17d5a67d4553d4","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"0a9cd32e9c0ca86a1f952cacc1f0a7916e48cf2af6d3a42bcbdea7660b9f1db2","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"120a26354a35b53fcbc89326cd074c241c3c9af2dc1685376a94d87faa460d2d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"9ef95d30f04739161c7a60f8c92a9b00c10f4dfb2a2d0af45b7a4cf389cfd654","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"a023a6271830323a492b204fa1e9d08e5e3bcf4a1948628d97bf2b0a194e921a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"5d9f512fef10ff55c9723a0722d4c0b2c8d4cdf78e10aa2e55c64670111e396b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"7a7d65284fcc393f4b2bfbe9bde0a9679c0aece3626096f721513884c7757b53","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"eed95c53400a8846ce126b0c3635033cc19b344119c2140963d14a193c456d06","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"649f1fe7e694bcd48e52b329eaec54861779d22d5fa9ca1f97cd2d0c15a66a9d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"fd523d2510a2bdc374d7e1c8e2eaf5b0a2f80daef20c7980530f6fd89c1bf7b8","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"e23df52066c9a715b1343206bc19849f7551faaab82a7196715f2892a4e661fa","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"f8dc276758deab74b2c76dac1e26d3df940121e5ec1128e853014a683562b5d5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"5b79e245474c4a002b0b10b95ecf11ca03f2af2cc588393b4b919197bfd2bd02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"8051426777397fc73cc99574a8e7b671c9170994a0695c154b2985f35b263ed8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"7f0154be5e2a1a542569140798e3fbe6ae6378c992e8a58bbe47252dbede87d4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"b3993d7dec40fbb05f87b44785aa166d36232c95fa22d2c403555ca7af3919fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"2b8c1f47fdc2b16f14e7b4e1eda803158023f4fdbae8a75b8c3acc22e4131db7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"52911fcd662da56ab42486e6960e1f1672432418fdef4303506c27aef31d2714","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"4076eaf99966bb8229403332a5d1fdf7b8254f002a14e708050f4f8eb220e8d4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"4f41e1adb44e1e6490f9d8b1ca8f082fbb073ad81f3d355b334979bc562c59ae","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"c3c0f3c543db4b8c6679837515c9028cc66e709a2e66bf097502c79660f85755","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"890f237d62e89fdf4ed0c320db8da076f713dcf086545f372846e08f9661a788","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"eac29f4280c08e41b08d5244f31c0be38573b0e1d4f151f36a558168f5627369","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"4b5df2dd18f53df6cf6a35467f24364b2ea597ae6a2e0517598190ad4efa48e3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"ab001b3d21210047e151947cbe3aad140e0b2318535b97ea05e4557d679fa195","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"fc7ba447bb5edd7f76d670055901afb4a628bb3eda3723a33401f5d3804be15c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"12eb7204c2912437adc61c876fd724ab152d1ba94b907dcc5324f8b9755ce726","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"1826fbc18fa62210bfdb8b15a583f76a6151995114768658e4fa39f1d6fbd7bf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"fd01e0ab0b85db461935a3618e9b459ee7a26eb57cb5cf7bccb187eb83091761","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"ec390cdcf9becae48217ecaa6fd4b29dfe14d790aca3a8b3e216d2bde6c10b4d","bytes":5310}},"requestSha256":"67ccfc539ec84ee81167c6d50750b0b3b40c71fb4de90e78560972781a72dc21","metricsSha256":"da14a579bb642f842840c1c64331a6146b9e8c49125ff21f893f8608e15fec54","continuationSha256":"76b7636452861e92b3b14970b3824bae909f1616b395a3c39f7bc9b5ed7756c4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-035.pt","sha256":"c3c0f3c543db4b8c6679837515c9028cc66e709a2e66bf097502c79660f85755","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-035.pt","sha256":"890f237d62e89fdf4ed0c320db8da076f713dcf086545f372846e08f9661a788","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-035.pt","sha256":"eac29f4280c08e41b08d5244f31c0be38573b0e1d4f151f36a558168f5627369","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-035.pt","sha256":"4b5df2dd18f53df6cf6a35467f24364b2ea597ae6a2e0517598190ad4efa48e3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-035.pt","sha256":"ab001b3d21210047e151947cbe3aad140e0b2318535b97ea05e4557d679fa195","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-035.pt","sha256":"fc7ba447bb5edd7f76d670055901afb4a628bb3eda3723a33401f5d3804be15c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-035.pt","sha256":"12eb7204c2912437adc61c876fd724ab152d1ba94b907dcc5324f8b9755ce726","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-035.pt","sha256":"1826fbc18fa62210bfdb8b15a583f76a6151995114768658e4fa39f1d6fbd7bf","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-035.pt","sha256":"fd01e0ab0b85db461935a3618e9b459ee7a26eb57cb5cf7bccb187eb83091761","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"289ee90a1a2ab898445eb1b3179e694ec0c5ce8d377a959e8e4c43a81478861e","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"efbc0fbecc6cecd34089980b75314ce6b374da0ae3e094366c22a621bcb9eb96","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d1835c98fe6ab1ae8a9112c94dffe867f135802ff1c1bb5f76d1d2a9820d4da0","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s43","horizon":40,"attempt":1,"job_id":"2481101","recipe_identity":"fddd30b411d0c9f1902a7d91d8d0e1eeecc9d195be7506545203d87876e9ea76","loss":3.4773626034714615,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"1155c6cfb11555ec0572702df6f13f256479ca34cffc37e68a2c91442883595b","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"1d574e9854cc86178a834acde9acc82c202ad989b426fd401661d31fe27acea2","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"f494489421d7013b2f1eef706036d5495a2d0f44c5cbea324c7621e769f9918c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"f7a4e835491dce247b50c8cb11390a2f7aff1f43ced3b8ba980275fb1ee7dc0f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"93ae8eeff3eb23f087b8d834df896923ddbd77b6713f14fd1bebd8a9c04ffa01","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"5fa6fbb31a623c7d8ae772eb642fc1f78eff4206e10cdd0de2d8354af067f3ae","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"91a60219321eb73d931bff11d527a3f1dc93dea71e1252b222626869548d22d2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"ad066435cf2b404d43ffadef0dd5516857c5dc80405e4e30ebef5f9e5864eb4b","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"3b26d8ff54f43a33b2458d6e2ad17e47efc78663276cd1ffff751e940b2d010f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"fefc9e438192955e194b0813f19a1aa1c80a4a56a963b96bf5672942c630dac7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"2e3bc316e02eb04111da4d50a1f2aae585b79f40cd76f0137dd39a02e05df09c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"a2b07bea8609ae0940fac4e88fc6549e9b2addf110f07149ba265496859ad81c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"d40a24233a9e7f20ee7ddb6dc318f37fa381c0c62fb89f84a9e3f88498775837","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"c12ca6e64ee9f4643e4e1cf1480a4452a955ddbbed4a50da588f9009b4a939bf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"3bc347cc3a7dbbfb114630fc0e5ba32e02726eab4b7e2af7b979c145d6d586b5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"cc7d5eebd4f09c90363f4ac16a0739dd8a923c8be1f394a1681e2ec0c3060cca","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"2dca0322b0f7860e36b37e882a1f665fd4bbd63fe3bc8187f74d1c4b7b185b95","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"2836bdabe814c297172b733f2c7d081b32ff132dd5e6d84d372784b7efbef1dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"2bce0298bd13de03d9be4a1f3561c11dee19f907c1483d94590c8ca2ad03be03","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"b1de50eb3a1a7235058497d66ee746b11c9069d96dad3cbf47d4c0e0dd8a8a75","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"8319f858936c41a9dfe28993ddbe3f785a26805c149cbab993e9e8f8421b2bcb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"33280b3d5940653d96f3c7fc0e1b43d158cd92294d40ba48d3f4bbe0f5fa7243","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"e896214c284b7058b9d067251b9c8f24b576394c33a9dfcd3b773f53cc12f75d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"05913ff574d25b08246d90a08f03b256ebefa3c2745b2929b0c36920a8b7b6bb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"8ddb4fb48ee91b94b240cab69c09bf71ab9843c02a5a8076c48741150fb77776","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"188a22a4a1437b9b29210bb1b1300ab107784da94ee1e47515d76c4a62072b6a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"38abd985381f12363149ceda401030c893db0d42015ec4add599338b21852216","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"98c17bff725baf9e7d08cb9124662b2ecf653286151013cfee47625236039850","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"cc28fe00800def3ce4979fe840f2e316b7e1c994bfd96ac9e197b8cd0e7829be","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"e39c46839fe2859dbfda9e27e576ea6b4c95595000713e7d6286cbd58f14c205","bytes":5255}},"requestSha256":"ddee3363dbabf1e8aaa1cd9d20ea3b6ab6a8fa2b1e7e817b105ac6a6d655542b","metricsSha256":"d18e9739ff023a16230a3e4d39563a0877c2affb224ab57529867c095a6fb922","continuationSha256":"d8f48aa6d376cb2079a50eaf97daf29e1169505b6b77444256d49efda5c511d0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-035.pt","sha256":"8319f858936c41a9dfe28993ddbe3f785a26805c149cbab993e9e8f8421b2bcb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-035.pt","sha256":"33280b3d5940653d96f3c7fc0e1b43d158cd92294d40ba48d3f4bbe0f5fa7243","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-035.pt","sha256":"e896214c284b7058b9d067251b9c8f24b576394c33a9dfcd3b773f53cc12f75d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-035.pt","sha256":"05913ff574d25b08246d90a08f03b256ebefa3c2745b2929b0c36920a8b7b6bb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-035.pt","sha256":"8ddb4fb48ee91b94b240cab69c09bf71ab9843c02a5a8076c48741150fb77776","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-035.pt","sha256":"188a22a4a1437b9b29210bb1b1300ab107784da94ee1e47515d76c4a62072b6a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-035.pt","sha256":"38abd985381f12363149ceda401030c893db0d42015ec4add599338b21852216","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-035.pt","sha256":"98c17bff725baf9e7d08cb9124662b2ecf653286151013cfee47625236039850","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-035.pt","sha256":"cc28fe00800def3ce4979fe840f2e316b7e1c994bfd96ac9e197b8cd0e7829be","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"1d1f8de56e4b858df1e95fc107b934d38a88d733a6d0ff0a9a8a18810c469dc7","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c40cb7a9f9ba296534c588c9cf0845876f79e406bb939c8f930d580c1452a7e9","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fddd30b411d0c9f1902a7d91d8d0e1eeecc9d195be7506545203d87876e9ea76","loader.blocks":797056}}},"awc8-wsd-h40--0.004--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s44","horizon":40,"attempt":1,"job_id":"2481102","recipe_identity":"5017ad899ec20cdea7253aed987bf7acd3424b65a75506480d52571ca4ac221a","loss":3.4839971104532155,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"d00e85f2f3fb0610c3b595169840a1cc9bc0d19910b02144a8fc4b0d6792a3c9","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"e55c4a3f5bf71ff05884525dda6285017f7b4fef67e59c8904884f92d676b8d3","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"84a95ae269c9b5b962c5fcb40930d2f52ef5551d5b825f857f0f80668da68a59","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"2dd14e9c2338c83a37f1c9b3739ddd6019af784314db09304e4aa0ef6a4cecca","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"0299e2c269fd3ab72699c07d7d8f748920c9cd3e1a163c1afd86bae1f791a785","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"63942ea8ebcb3316c6f465c5bd63a9bd8e852ad53782783b6d4b01bd33f496f5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"0e7f3f5fe843724e9278dc645cd6f0d8a6299dc2aef77fcfaf7cc8e919ec6747","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"3edb97239cfabf0363204469faba93f49ffac16c15f878d9dcc7fa7168eda972","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"845abc5e1bd366e4e752c851f9fadcb90a0af54b0d4301f8f36ff343b03753cd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"faa21d37947b151ed8443387d3d15612d76f651c7313d75d12fdb561c293135f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"4d8e20f0b7143b4985ca42155c8e4ca05dd6d359ec581c13993b76a3acde5fbb","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"edd734d87140941b2e2520ab021751adad26203f4dabb84a2d82cf8bdf6ad48d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"c7e412dcae5342e6c12d41072d443d45847c4552ad4fc8ba870f6cb8ce66ba2c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"d7a36144eaeba630f0f3eb657be3e03022e5da2224c4fa3a3aa0c87c73db270b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"5c335724f9af4a0b51fe19e8d5df24ddd3a7755b8fad022142edbbbdd5d1ae7e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"1e5ba30ad70697e5cd1c4626d3e753fa26ede5f6aab300ab65ea01cfaff3804d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"98a67a787fcc7e527a2e369c7827cce51d6cc9225f040feb6fd89e0919254bd3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"e1152bc8c409c7bfeb8b224eed60fc0ff98e8fffe8e7cc4fb19c0750ae04af22","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"23691ce948d68b38e2e6f5214d9c5e22a844e9752606c53311f0b359558bdd67","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"50347736fb52a96552f745692f27511267b8e651043b75ae1ee5cd33780dd4f6","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"d06626565c613890b416479d9dfde00a07e671ecab5fea897679964b2ebd5b1d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"b749cfe0ce86229dd59d2a65966af74aae873e116d40cf520f2991aa29ef8f3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"8063f1f3483743e474659f344a4fa7807e4bfca34fcca0f722b6f783ffee976d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"9e39cde1a37bc83629c196db9d5746517f6a69d43a85415e5f552793abaa3b45","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"a6338d283902b413e87f6b570c7081f3a84507ae9240c4451b1194451542ac40","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc882d342bba603fc791d0a283497e1ff205153ad4564b266d78f05e27e50808","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"c3b0ab747902f16a150a3ec919b9fbac9a009bd97d2c45850a647602add9faed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"2f8c8b394d107169374d621ac9a48928cd2edc085317d8995fd0e50853c16247","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"3edf73d80b8ac4d926c1bd2d27f19c7a70d3bbbe2e2bf0051b8776d597c6c631","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"5d209ef02fffafe2327f5c2c269ffa2582e16c0c43c9c247f3cb617a0c774eb4","bytes":5309}},"requestSha256":"a420209d9320b5463cb1ed4fcb9298cfdfc79bb6d35459717a31c8e79c9d0526","metricsSha256":"3c0c4031ae0dc3a2edfcea3afd7d3864d98ecf21ff1d1e0dbdaae79cfd28cf39","continuationSha256":"fc117883a71f86cecd6fc989f50fa262de5cf00f1751fa037c6dc64a2b206f7a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-035.pt","sha256":"d06626565c613890b416479d9dfde00a07e671ecab5fea897679964b2ebd5b1d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-035.pt","sha256":"b749cfe0ce86229dd59d2a65966af74aae873e116d40cf520f2991aa29ef8f3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-035.pt","sha256":"8063f1f3483743e474659f344a4fa7807e4bfca34fcca0f722b6f783ffee976d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-035.pt","sha256":"9e39cde1a37bc83629c196db9d5746517f6a69d43a85415e5f552793abaa3b45","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-035.pt","sha256":"a6338d283902b413e87f6b570c7081f3a84507ae9240c4451b1194451542ac40","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-035.pt","sha256":"fc882d342bba603fc791d0a283497e1ff205153ad4564b266d78f05e27e50808","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-035.pt","sha256":"c3b0ab747902f16a150a3ec919b9fbac9a009bd97d2c45850a647602add9faed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-035.pt","sha256":"2f8c8b394d107169374d621ac9a48928cd2edc085317d8995fd0e50853c16247","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-035.pt","sha256":"3edf73d80b8ac4d926c1bd2d27f19c7a70d3bbbe2e2bf0051b8776d597c6c631","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"a573079db1715588cf2c504bac9aeb47137c47a2f1b1a363018dc80f6ab7e7b5","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c060d02fb5238057d2ab4eec69a11852b15b5fe6e2de99f36ea8c876fe5f104c","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5017ad899ec20cdea7253aed987bf7acd3424b65a75506480d52571ca4ac221a","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":40,"attempt":1,"job_id":"2481103","recipe_identity":"ab277b0ae79c238e590aaee645ff1ed1a6de612cad610d4c4cced23aef522150","loss":3.4750812273482987,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"7bdc4ded78495c557bcd1257496e5fdbc75ef9f399b006f085e51bdd241c115b","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"b4941a2a103b7affda4d6191506cfa88b202b9e68f64611c837a35dc56b7fb39","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"639111658f0bb8dfbb6cf643b74f2923d655452abaf14128defe3e73017d45a5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"3a1eb2205afe69c38ac34e4357f463b40d78622f1a76d1b7d0f3675602ecee98","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"4d17b255e23227cd65b3175e576530c7238478ae77857a5ed6bb88f3d85f2f4c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"34cac86b7b3a1884c1719b4c1f7742b0e9aa774c1a123b587472c43108dd7067","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"844dff12f73eb75a0c63e9161632af43cdd0bf302507e438ae372e8da85534d8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"15568d13dc38331ff5a2dbc5a8cfced300eea32e2d68f972d996d54212acd35d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"1cb2dad36752baf16e9a01122082f73b5e78d5c4eb5f8545de881f638365a744","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"5ceaada0716c4a3739940e3346fe93274f7d9912c8daf6cdad931b5cdf256e73","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"7d63dbed768166ef17bcf6314aae5bc5782f4096a94b0cbe5aa1a228573038f1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"fe20269a017cf7e9f87d29133253f946a20caa8c69702e8bc2a4c8442c57f483","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"f0462d1061975650b8add703ada21a8fdb8c0b4c6d01bf67acb38a040aba0c08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"513b8ba32b38ee07c4956329524a86c1d0b2578b5dc644d19d898fb1e5047934","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"b6fde3581cb43cf0166ff9ddbfc18a58c24a2a8d1ca8aa58354159cc88cf0757","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"7d41bdc1688359ff39a5656794b36bd7ae46af856c527ce811989a7b3fe7ee07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"e8a0c8df455031261b40f8e3c344d4fda5fa865c3c72f7ae706bc4823fa082e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"2fa1ddbffbac321d585effbee3b750736892fa5f0195f8198bd32098101171c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"3bed072c5ad8337986d8a75cc2229c2a80dcf98d2625e7e270488ee63d8d22a8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"a5520f79c639c87ddf2cd83a9a5c0af5b939f4be06bd27d18a6d6939a547cf40","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"f73a5f2e2144ce0daf03ab63017ab1d0a8b13c54e710bb6df7980f1d66c28845","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"bd2dbfd04c4b333df0ac0134340d3d1685a72cc48c5e3ffca081c74dff1147df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"64db6878edb2692c4ab80c4e94aed665eba142c19945979d809624bcbcfa3a6a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"9d77ff8992630ae47038d52493a4a0daa35d329067430dcb112afa233a29e86d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"1c1e36ee92dda312cb8d21a3009a0f6b33ea8eca4549fbb9d695a58d8c314d72","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"4a2c295939088317dbb174026b89aa70ed7985bad37fbdc20a3ba7a5b12c8564","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"d79a4ef2afc5f2d4d357ba540f00aab845d69505ea2229a727f9d1ba02fd69b6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"6c74bf3be268f1c170b038ff60012c4e29bd94a531483d2649d50601bbf5fd09","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"7e84212c0f453b02e8bdabb235224e5cb21da6aa47bb6bae4d8c1eec2cb6c42c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"811b57a648804e4939ff5e086aaa4bfe066604db960e148cacc868b75151f802","bytes":5312}},"requestSha256":"aefc5a38969d7a0620d20ffd3f22a8bec3572e8b646f9f9a25f4957f51bc80eb","metricsSha256":"6e6d0eddd4ea11caea795c8b6ac38f489fb82d7c5c235ce22ffe889766b0e6a8","continuationSha256":"f97effd4c20506971369e884cb6c22810d114e95e5cd562658a917af0e90b849","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-035.pt","sha256":"f73a5f2e2144ce0daf03ab63017ab1d0a8b13c54e710bb6df7980f1d66c28845","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-035.pt","sha256":"bd2dbfd04c4b333df0ac0134340d3d1685a72cc48c5e3ffca081c74dff1147df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-035.pt","sha256":"64db6878edb2692c4ab80c4e94aed665eba142c19945979d809624bcbcfa3a6a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-035.pt","sha256":"9d77ff8992630ae47038d52493a4a0daa35d329067430dcb112afa233a29e86d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-035.pt","sha256":"1c1e36ee92dda312cb8d21a3009a0f6b33ea8eca4549fbb9d695a58d8c314d72","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-035.pt","sha256":"4a2c295939088317dbb174026b89aa70ed7985bad37fbdc20a3ba7a5b12c8564","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-035.pt","sha256":"d79a4ef2afc5f2d4d357ba540f00aab845d69505ea2229a727f9d1ba02fd69b6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-035.pt","sha256":"6c74bf3be268f1c170b038ff60012c4e29bd94a531483d2649d50601bbf5fd09","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-035.pt","sha256":"7e84212c0f453b02e8bdabb235224e5cb21da6aa47bb6bae4d8c1eec2cb6c42c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"cdb5f5685e31ddbb323504f49800a1fd0c8fcc055d32d92d4ded25eef173132f","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c2692992567ea6b3fa916bc28c202b0d89d28e3579cc707c1695d6b5260a3a30","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ab277b0ae79c238e590aaee645ff1ed1a6de612cad610d4c4cced23aef522150","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":40,"attempt":1,"job_id":"2481104","recipe_identity":"916802bdb54c05fdf34bce89371003ba29b1ab7dbbe16e56946184db4bd70b39","loss":3.4859453615064666,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"f5979b8cf37d58aef5ce8301671390a07847b78bac0d2a2281d2fed11c2af767","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"b86433286d3f753d3f1232cb8f73ad197a0bed678bf0820f5636da2a79ad6b23","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"c5da851676cfb45b0299a25fc0b79ea60b864590bb2206e5c7bc8ca1f6259380","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"244264fe04578307e619fd6d67bd47f84229cd0616a71335b8f8380d44214c92","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"824911a95083b522c71a7921281865ae8905a3e5eb11da74c3ede914bf344422","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"5db1f3f3a9f2ae300d0a624d8984decb1d2110fd187b1d37ba22d2f55ecb4c54","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"456803065a6ccab115f89964f1527882ed632f8d30154c5db311d1fef93c5ba2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"3c15021dd8bf089fe53bdeb4169633d92a2f5194d5e724d5c3e7115fa20d5251","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"3b6796ec06c4179971ca8fbbfbfedce6cb6bbb5958b43884cebeaf55737327dc","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"70898c4b02fddb880ee1266793eee8c7a82cb961deac3c6a1575fe0d623893f4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"21ef0c27a9c1e38e6cab80fb678af3c013a8d3449861f71b2ee4041f086047bf","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"4d3d69ffe618f7f4d525bb10169cb0fa7d07c77f3e5cbe5e2efb0b2aef049171","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"fa4a2eea1e4b3f8542e77ff42af08d2f8edd26cb2bebf278d3d703e6c7bcfe08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"04bee43a1e3d732b9ebf46c85e3dc2ad79da192daef539feaecd77bb6f37d1d9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"69748e30db69badf9226303a6da3207869c10a555e691f19de030a8433756404","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"3fe530e559156c6f78ceade1b6d8ae07cbdf9b489431cc9d8f67f4e5a7af819e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"6a6456eff7d372823708cf2497bdde29dc2943c487a814cc6ddcf6d9679f82b8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"7cc6c5c5dd952ea3637b8609dbdc30aa407d73c708ce6260fa5e98b97fb036e1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"c41f8136b858fa0a7dffe9f2e5aae08ba14f95f7fda8c26f202c347fa5f241fe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"b52320d6148e0255efb57dd72b320866919e411fc5558114e601c59e0c32cdaf","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"129d2ceb02f75a495ec1303492c549ea4427760a30353e6c1a838ee90203b0fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"650690386e6081a6edeb3ab8814be8b3e978e78697fba580b58155fab956864b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"68b9286b25ab838205587b2162aee2189880b17f0ec6d52f03f8c19bf30eac87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"733bbeea6cf124031f930c4e2dfffa212154c238eb987806cb0ac81a18e7e56e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"796c17e93a537b9ab75525278fa2204357dd344a547851a9b7ba700d6c299f09","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"768bea380d3c7eac92f816bbb09e332f2e175d8f51abaa4df57f87db8cdfcd56","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"9890ef791815e67e274645257c7c130533939ac8d8e8a927d925c3e13027320c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"65397ee439f3603e5e2cbe54ad35be996f991ccdd291ea17a2419d99a2742af2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"62769910541261c88156abc70005e30d1c84c02f118864b27bcbf10df0f60dc1","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"1b40ab67d086a1678c7302ab38f83cbf231e70262cc176bf0d1b3763ab9fe582","bytes":5307}},"requestSha256":"e50e05c48d73235e57fb2c511c84a31a5d893bd70a8b4e4b9fce7cd197d4770b","metricsSha256":"55c9785e28c19093ff69b37f4673eb2c6d41e63cb32cd3b2b76a1ccb02833002","continuationSha256":"fbdb36e407be631d5cf115ccb21e94f47960c4fbd454cb82d0dd3f5d789dec92","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-035.pt","sha256":"129d2ceb02f75a495ec1303492c549ea4427760a30353e6c1a838ee90203b0fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-035.pt","sha256":"650690386e6081a6edeb3ab8814be8b3e978e78697fba580b58155fab956864b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-035.pt","sha256":"68b9286b25ab838205587b2162aee2189880b17f0ec6d52f03f8c19bf30eac87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-035.pt","sha256":"733bbeea6cf124031f930c4e2dfffa212154c238eb987806cb0ac81a18e7e56e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-035.pt","sha256":"796c17e93a537b9ab75525278fa2204357dd344a547851a9b7ba700d6c299f09","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-035.pt","sha256":"768bea380d3c7eac92f816bbb09e332f2e175d8f51abaa4df57f87db8cdfcd56","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-035.pt","sha256":"9890ef791815e67e274645257c7c130533939ac8d8e8a927d925c3e13027320c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-035.pt","sha256":"65397ee439f3603e5e2cbe54ad35be996f991ccdd291ea17a2419d99a2742af2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-035.pt","sha256":"62769910541261c88156abc70005e30d1c84c02f118864b27bcbf10df0f60dc1","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"694ea975a6ed8d98f70e0c54127d147b3f12d7d1bba53dbd183184089c0051db","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"387cfe8165347ac163a74a0316325cee9154bebdc8446e2f7125f5e0da24ba97","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"916802bdb54c05fdf34bce89371003ba29b1ab7dbbe16e56946184db4bd70b39","loader.blocks":797056}}},"awc8-wsd-h40--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":40,"attempt":1,"job_id":"2481105","recipe_identity":"c844d0a29cea7542c33f01c70c5eaa88d94f47b11345237b6579635b556d590c","loss":3.482726184281099,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"3257769c19567d333e262d1f146181972471af83e248b0536b6e0c8054c2b82e","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"6e864bb687f04c1bc667e20977e0d9d92d04e5ab1558a4f8f4d272878d2e8e05","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"04aec635f0beed9c4dce895923f9110805cec655fe501284afeb3761524ebe68","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"38c5061bddafada70ee3a487f39de11379b8e01f02563f78502064d7a32097c3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"8f50e7381d77e02182203b52eab4f6fb2d5177244851ab2a0b511424f0961f55","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"a831327083a36b1487fddd7f93bfd423cc2bc11b0b56578084e085543df527a5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"bfd0349f84a43a5308ac3c469b28332097ddda75bcd219bc6d585bd0deb53f7f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"75c3fba69f9d0eb091f022be9fa42fc683e9bae8fe1a992b01268adec53c92d7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"15e6b59b52a116dda9fe3d39e2869311c74e590a0593e8ff28ede90f6cbb6117","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"8aa61bd48787f539cfd17ad962b8b50fb61a0e000ca8dcb0fccf73a45f457e10","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"e149eda4a0356edff71830faa86cba1f052a3dd6e8e9d2182ef89652619cc419","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"d8e8c4bd87e290188245252a082b38377977b7008f6aafcc08285191d0469169","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"ba269c7d1a3e6fd6eebce28ae61d5be8994cb4d4459893a973abdf992d7223e1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"443b3ff78606bbb16da279ff2da0131ce0a729c5afcc7d9c242c15e7ea25fad9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"355c486cd54dbcdcdc6586072df4b3926de6c7396747e28d49a44862ffcd03d9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"0456208e08aa1261bcb43ca3ad14ec0ff45f45aa244b74f8e9a2546604c27f01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"6dc3dc8a1e0351dcff3947e88328862733d47380bbe51254e591d947593da2f6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"49e6621b4600f0f59ac11fb6f610a5b793f7966d50951a8f7ab14c805caa9a25","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"ce280d34f6074ca401024ca0b67810ec316dd8e9848ec08de94ef87a3149b22e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"032aff9aae451f6634d6fc9cefb593507dea1ad07cc0cddbc8a1d1bbaf91dedd","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"87ae862144b80158547657bf3f2e51f719b081a71fe040f3f6bcf7f40f2799ea","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"1c63e408bf41af32284de7945ea776fadb72497830a356651d6f2e80bbb22c19","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"54b7153daf066fae12fc801004ad4746a95f45d8d06ce7e5d6e2a86e2fdf3ed1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"089be6fff7b4ff85b5a5d75ed1ce8bb85ada215ea7f01844bb7d560c7fc00143","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"52962afded4ca1a0a947beacfa9cba20e3d3263b47b3a4265abf8261d7dc6c9b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"20ea1fc12d452737059de0a5967c3c8d887e4f1294834fa6b8819cde6f56d563","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"7b039bf20a76248d100f6190e9d315f41c43303fa8dc95e69354eddd3ca953fa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a490cfde6a1f8ba1c7c8da4a5a970640d466c6c004422c8e392464384c6ac2bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"406bfc3340154c95b9cd832ace6577a8336a5e601fd5534b5700c722f6c3c7e4","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"fa16a75bc386af0f5d75f3207b96451b961b568692251730a60cddbe029ed590","bytes":5305}},"requestSha256":"409c4870b65f4fe7db7dc4dd9fe7ae40a03f4f732ecb7c8a457ee19f8e8c6a81","metricsSha256":"dde056173320cd2baf1e92367e4ada8073093fd062f37fbff10ef2bd470e43e4","continuationSha256":"5abda6947951e99c652be1b436a984653a873777e5ef1d1bdc221ef892ade9a5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-035.pt","sha256":"87ae862144b80158547657bf3f2e51f719b081a71fe040f3f6bcf7f40f2799ea","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-035.pt","sha256":"1c63e408bf41af32284de7945ea776fadb72497830a356651d6f2e80bbb22c19","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-035.pt","sha256":"54b7153daf066fae12fc801004ad4746a95f45d8d06ce7e5d6e2a86e2fdf3ed1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-035.pt","sha256":"089be6fff7b4ff85b5a5d75ed1ce8bb85ada215ea7f01844bb7d560c7fc00143","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-035.pt","sha256":"52962afded4ca1a0a947beacfa9cba20e3d3263b47b3a4265abf8261d7dc6c9b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-035.pt","sha256":"20ea1fc12d452737059de0a5967c3c8d887e4f1294834fa6b8819cde6f56d563","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-035.pt","sha256":"7b039bf20a76248d100f6190e9d315f41c43303fa8dc95e69354eddd3ca953fa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-035.pt","sha256":"a490cfde6a1f8ba1c7c8da4a5a970640d466c6c004422c8e392464384c6ac2bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-035.pt","sha256":"406bfc3340154c95b9cd832ace6577a8336a5e601fd5534b5700c722f6c3c7e4","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"8085139d3dad5eadc7c23e3aceecb2c6432e0588ffb2fba5cdbde238c66ccefd","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dba98e8ef258152b440ef9af80a2e12411c8eaeaff62703246ec15bbac7e2f15","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c844d0a29cea7542c33f01c70c5eaa88d94f47b11345237b6579635b556d590c","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":40,"attempt":1,"job_id":"2481106","recipe_identity":"095acec6c28d4db570adbf8aafc6d3ec619a58c775cae51c5d7c865de6cd51bd","loss":3.5072350134506016,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"5f9f7e295788b67a95365c8043df4f56b8cabbc2c302ee6c0c77fcc1ec4f1ac8","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"85d3cd6e902a471bdae9cc2d3dcf7193875b79a98accc45e081826e137b52aa3","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"32a578e0a638de03c211f56e89a36b64cee6e851334abb9c29e0d0332596d4ce","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"06bf0aedcdf434d400a8ff332e67e74c6f5e49a04d6a85784385770ce61652b2","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"171cb35506d0c6c9f402154f554ffc5ed458974e91f5a61c4709ed24c7aac38b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"4944cda012db815be3d3391599567bdd62ad767ec6aaabebb2ef0bcd6fccc887","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"e9d17221762bb6c4d396bcd3a2c06c4ff33377aacc79e3825a40dd5df1ae1b33","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"24bf826ff92a54b3ed393fe409db798aee1fd829036834d06fb5edfd1be4d12f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"caf70aaa1939f51f85151ca271843862b7f58b39ff0c3c7be6ec034e7bfc8303","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"ee9ab64911a32e06cc89bbc1c028fbc59330c70ec15089cb1e67286d552a45f5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"95347f7da0eec4c212f3a5d4531818f1cd95ccfef53c5b007b7ccabf15ba08d6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"3514c7854042c19ffec15d8e4b6b2683a9b2f633b06123f3486b7e66ec6ad5e9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"dbcf3cd99798eeebb7018961139fcc4219ca29617afe26847aa2b9ef4947ccd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"33d3430afe67735f94d28a7f228ed10eb4b18b6c94042929843e01a448a8e9cd","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"750bcc37fea1b026d9d057ebcd22fc6e1467fc742201e14f47c3b001fc09d5d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"238c711be143491b4e3b6027e630ff1797d5619e467d8ff03868c04d9ec7016d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"c09dbfdc7f788c71733173c22c47db194a7f784aa6e15e6b918d8b23c7a79a9d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"945f5714f43c76d3ff4019f0fac54421586bc0182419473bce513b4346907ee2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"34a09e81f768263acc603fa64dd5f5cf2fbcb68fdb43b425caaababcb9f40878","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"990ebba1094b5147a50cf95b3d804cbd4bc77c36ea1306237c9f585e70935f32","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"b530cd36f8de24ed50ab6206f85dfcd930ae1847b1e8da4050434857876cb085","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"a3aa32db6ec00877328db1a520010a34441c66fd9a88bd125a82e05acff1be4b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"a3dd1453259ae08b189ac0b135fcdf5ce6ea876dcae10a7f30f06b346310a69b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"2322ca0cca3b2b3c3171a6a988763125a2e2ea9f18137171efb54fc25f416daf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"30000f5926c32549bea14abfaea9028544e8b8ae3d7c80dd1419796c733150a7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"3549e57fc3ed8a5d831a6c9a791626a52b0f9f4b6143024de849820c4b323c0e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"5f2c1a0aa6289a7786f798cf55aae2d6a6c79b54724cb33870853f141b8fa351","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"f93fc4234d77448c8c8b72f7146c22dbdf4b656d9094d0399c5db97a7ad31e91","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"023108c90068b41af9bb580cfdaa75450fe62fd8a97996715b9f11bd6dd2339c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"2ef9c9881ce6e8e42ae1f180f27886dfbb611c83ea04b083419584bdcae006b3","bytes":5248}},"requestSha256":"6426862e1abeccb8400adc2b192d79b146b6342a99a8ffeafc36857c572e631b","metricsSha256":"9b0ac4caa9bb55400a4daf25439be1b1357ceb06492699510674550855afc6ff","continuationSha256":"2cf86662a33956d3c71efaafe6a048253690bb397caf40e0c7f46be94003a359","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-035.pt","sha256":"b530cd36f8de24ed50ab6206f85dfcd930ae1847b1e8da4050434857876cb085","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-035.pt","sha256":"a3aa32db6ec00877328db1a520010a34441c66fd9a88bd125a82e05acff1be4b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-035.pt","sha256":"a3dd1453259ae08b189ac0b135fcdf5ce6ea876dcae10a7f30f06b346310a69b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-035.pt","sha256":"2322ca0cca3b2b3c3171a6a988763125a2e2ea9f18137171efb54fc25f416daf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-035.pt","sha256":"30000f5926c32549bea14abfaea9028544e8b8ae3d7c80dd1419796c733150a7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-035.pt","sha256":"3549e57fc3ed8a5d831a6c9a791626a52b0f9f4b6143024de849820c4b323c0e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-035.pt","sha256":"5f2c1a0aa6289a7786f798cf55aae2d6a6c79b54724cb33870853f141b8fa351","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-035.pt","sha256":"f93fc4234d77448c8c8b72f7146c22dbdf4b656d9094d0399c5db97a7ad31e91","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-035.pt","sha256":"023108c90068b41af9bb580cfdaa75450fe62fd8a97996715b9f11bd6dd2339c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"0fd6a044e3c10ae37e562f1e3011782ceeb1e64809b9222603f506a27be908c2","bytes":1958918101},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f810df0244f56fd9aae0688279ff5949bd8ee7595b5e82a82ff4d739136dbe01","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"095acec6c28d4db570adbf8aafc6d3ec619a58c775cae51c5d7c865de6cd51bd","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":40,"attempt":1,"job_id":"2481107","recipe_identity":"e9ee41b7330175910f22cf2570f2f3bff8208bd387bf18336e9456431a44c54e","loss":3.5055125890623753,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"c41087a5d4f45188556c3cde36de3c6bf724a1f3ad002d542e6e160111af8ec9","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"974838c9c67fe6f7b812fad8dbc49e8c2665b48da8daee55424a38adb18275f8","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"dd09edc156995df8452470ceb1db0c55878614bf2706c4b2d394593a68cedd0c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"ce25a82c03cd37122bfda14ce8f88e2a3cc4a49f1bb43fd83f9a796ff3f9c0c8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"ff6dbd44553e1f30eeffb948db10e4349867c3c13b05f983af96b11343be8636","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"6df2deb1ade79194ffb83aee7d95ea3434f4e54d6f741cece9742ecf62ed5b37","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"0c8ea29724eaf0932474d00490eb0b50afcad8abcce73a3594df5288917f6368","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"986d7e7083948c081d2cdb88878d35643e406c946a0fa72e70b3e239555bacda","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"a221fc6062852265c01ea3364d7231fdea8a3889396b097796479770ebd58c26","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"e45f25f3ece8935d99e0395028b7af4958e84d6828a7a90ea23801466447cfd0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"b30e84ed50634ad44e1e7ec733033ae81a57db5896c4bc279c8a42ef076f2858","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"8d82d8df2c16a4aad045d3ab6b1ed8afd2064fef7635279922930855abe1fa7a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"cf82cc0427952d9cecf9abdcc697185ff56ad0f3d74b478f8f9e5c015dab45af","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"3a440ff76cf55450df1371dc3facfd2dc1c44041a2f0ba590c58f461ff82c0ee","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"13392f95a92d6f5d9978159c4a8ab73ebdce24790011dacc2385a7d9ddc6f4b1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"ab6eb43014b2d33a5e224f8f6cb62e2598a970f87d11ae5d1acdddd5d3a2ef16","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"fcb13d2a20f351bf409bd6c813f6152842c1a83343a640171c4cedfa62b15d92","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"9cdf99e417f3a243ac28e7f3f7d023f064b4cfec5950c181e562e28b0cbbb8ad","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"9cffa9f4391746cab1187217d451e9fcb22bb07d352bfa712569dd5e9d056451","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"2d3329d8ea65b9809b9e19d17ae3d17c370805ae74ecd4e6027d63b41effdb55","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"5a10c222624f919352caf2c5954e1510ceaba6fd39d42e23d0b1d56f156bb8e7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"1a0243c1b5076ad55d51be62443edf2c3af9700451bb2096f05b2c86f8bdaa36","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"7571e66ea5124b1ae8a7abfb32583567612698ae546a40bf62a1efa22fa20fbf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"4083ed3d6ae100a0eb494e3eeac92db342dd9c9494138d468965ef9270406f1f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"b1cd3d0a1ddba2b4625445dfb43f82a22d0f001b076b8d7a2ad9da727c688cf1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"73507de22070b4767700c643d89e7b53cafd1f4328a715e5e12b8c1a0437297d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"154fef8ce2434de3f7685a2ec4fcd7a7c42ee3f3a731e79044212f0e220a6707","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"d6b385acf7f8088246d6e334a7339bb43e1640c826961b829a08ceb9ae7389a3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"78d575288b63f81a0fda1a84a29b1fd08afdf7e38e4b1e9af5db15e60871ca4f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"6c708b082cb9e1f76ce3ac2cc9e74e7cf0936fca40a99a7f0d4c0c111a9f1e95","bytes":5255}},"requestSha256":"0e15c657245b0d4e79b5497f1b204314122444b92f58463a5079381fb731ba9a","metricsSha256":"a83c8ccb5255f595e9dc5b5255808814aec972014ed2ad744d8bf5f71301c587","continuationSha256":"c036776a7b5419b47f24d78f609805eff826e3398a4a6350103822968e8d9c4a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-035.pt","sha256":"5a10c222624f919352caf2c5954e1510ceaba6fd39d42e23d0b1d56f156bb8e7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-035.pt","sha256":"1a0243c1b5076ad55d51be62443edf2c3af9700451bb2096f05b2c86f8bdaa36","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-035.pt","sha256":"7571e66ea5124b1ae8a7abfb32583567612698ae546a40bf62a1efa22fa20fbf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-035.pt","sha256":"4083ed3d6ae100a0eb494e3eeac92db342dd9c9494138d468965ef9270406f1f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-035.pt","sha256":"b1cd3d0a1ddba2b4625445dfb43f82a22d0f001b076b8d7a2ad9da727c688cf1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-035.pt","sha256":"73507de22070b4767700c643d89e7b53cafd1f4328a715e5e12b8c1a0437297d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-035.pt","sha256":"154fef8ce2434de3f7685a2ec4fcd7a7c42ee3f3a731e79044212f0e220a6707","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-035.pt","sha256":"d6b385acf7f8088246d6e334a7339bb43e1640c826961b829a08ceb9ae7389a3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-035.pt","sha256":"78d575288b63f81a0fda1a84a29b1fd08afdf7e38e4b1e9af5db15e60871ca4f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"1f13b7397675694564c91fb10979bcb287b136cbf1c400739d09a94f6333e6f9","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f4e2c782f1bdceb4fdd606b0dc0f5506e7d46d28d7d048898a248bcbb5e6daca","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e9ee41b7330175910f22cf2570f2f3bff8208bd387bf18336e9456431a44c54e","loader.blocks":797056}}},"awc8-wsd-h40--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":40,"attempt":1,"job_id":"2481108","recipe_identity":"a365bf3a61e7c80eab29f61e0e52cdf05644b0a835a131d7c459ea03e5ddfce5","loss":3.506473711511068,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"4b73258af7afa0af8bf336572cd0ad8aeffe978668c7434daa34e3123be6bab4","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"cc3328d1de299ac7ba82f7569de851ae8669c100376dd9995f1f8b4906e04d74","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"1770b4546bd3961bc21d1a52fefec45f817bec3911532195a5e72f3205ce8702","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"9bd56566c6ffdc01a16c5652ba9c0356051d171f8376060b942fcf27cf51f0a7","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"3a38612f66d1e80a14bdef1b9ae5ac4b34854047fc636d0bfb02a50f0a4044d7","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"a9e53240536a4e418a40b8585cd1d57a200002da39d5aac9ebd50b2aaac3ee79","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"53da7d0b3db00a535952ad18796db232cdc8e6c156c451220f850e9e956c12ca","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"3fce79d5ec91be4b8a5ad3f171d553e18b537f6d4e0b681c0793fc2ae1470775","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"3f1f0fbe5555af7988e67aa5d73a3bd719f1c28abccff5163c634479bafeb3f8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"9e84374a865f29f523c884ba4006ab8edf5642f00eb4167707f5d60b14c43625","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"d25b8880ec76e656cd6f856ee034a913410edba0c060ae679d09c2cf1c5d3fca","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"1eed95c6f0950087a330412d35d6e6cbc68b118e439357d406e01ea4e515760e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"e4bccb31b73ec04499e016eef5770ede0386542add482d6350fdf4a5d495ca48","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"284261ba99aa2506c3f20754dc58b27c4f07ab0e845d4bfd3cd17554e569677a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"e06b78ca36b2e054a01550df7b0a30e072255b58bc8930b79b79fe61a99ffb49","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"14327ec46293fd6b0233201b2384f720b7d68bb463c4bf859316e6576740dd69","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"30d82771a55cc2bf03ca7d6df97a21a38e53c015a2aafd0457fe326f38828436","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"2e45cee83ba937594b899c174a35ea1d8d9c93c3104b8fefa6f5eaf523ca4d07","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"f1f2ec21b4936e6148e4f1956e6e9a1e1b1db0b4515d9a57902c60896e269eec","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"8ee271ec3be83c154f9d3ae050bebd410b4b68a0c71b744c86c87d0a5dc352ff","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"21f6d926b27950e1ac2dab4ae1aa70db6396a3aac040dc0d552e5ef60b43b53e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"02cfc3054480e807c5a6d646a1a52fc3c201d0e665fc446aa72f48b04c66165f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"ab0daff2844253c486411063a50fdc243020a4e4ceecf3e1a10892fc321354f9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"7b01d563655c4163fb1f10211844ee1d73c906ed534fc894eaed24664bbe280f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"04e6f01899d65e6e23141b33045bd36a9002c64a7b0225906dded750695d2538","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"975aa2c3e936dd6fe5f036b58b2605484d263a348dfd3756d374927751160038","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"a583209060ee79a24dede23ec82e5b2e5defbe60d1e2ad1fb0a294508d20b755","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"e80116c26109538136153681d00edcd22114dfc8df45e62919cd0d93f434adc2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"a73e85b10df2ff680604ef211baee647e57fb6e4ab6bd921791897247ff6f7da","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"b4bf6a80e3619dc1abb0da8f7161d16df7285a950b825e77135f0bfd9e6d5814","bytes":5341}},"requestSha256":"5d1daf3a8d452271bfb77d1f1eeb74fe729ff7c912f716c8bf1b26feba1d2f79","metricsSha256":"ebc04828fec5dc38b6fb500126270d677ba4205d6efbd522f884b99465be27c3","continuationSha256":"6e1b3aa77c4d484b2f51c47d4690a9eb24f23ed5610595068b92ca2f2a06f565","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-035.pt","sha256":"21f6d926b27950e1ac2dab4ae1aa70db6396a3aac040dc0d552e5ef60b43b53e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-035.pt","sha256":"02cfc3054480e807c5a6d646a1a52fc3c201d0e665fc446aa72f48b04c66165f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-035.pt","sha256":"ab0daff2844253c486411063a50fdc243020a4e4ceecf3e1a10892fc321354f9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-035.pt","sha256":"7b01d563655c4163fb1f10211844ee1d73c906ed534fc894eaed24664bbe280f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-035.pt","sha256":"04e6f01899d65e6e23141b33045bd36a9002c64a7b0225906dded750695d2538","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-035.pt","sha256":"975aa2c3e936dd6fe5f036b58b2605484d263a348dfd3756d374927751160038","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-035.pt","sha256":"a583209060ee79a24dede23ec82e5b2e5defbe60d1e2ad1fb0a294508d20b755","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-035.pt","sha256":"e80116c26109538136153681d00edcd22114dfc8df45e62919cd0d93f434adc2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-035.pt","sha256":"a73e85b10df2ff680604ef211baee647e57fb6e4ab6bd921791897247ff6f7da","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"3025eae0a595d2f87bd982eda4914fed4d29295340bb4fb58873e75fdabaa6df","bytes":1958918037},"source_horizon":20.0,"target_horizon":40.0,"epoch":35,"step":2724,"cursor":348672,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":20.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[35],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-20/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1b7211e3f29624e396b9c48fd1875bfabb6286cfd260c8af322367fd030273dd","loader.blocks":398464},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a365bf3a61e7c80eab29f61e0e52cdf05644b0a835a131d7c459ea03e5ddfce5","loader.blocks":797056}}},"awc8-wsd-h80--0.006--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s42","horizon":80,"attempt":1,"job_id":"2481774","recipe_identity":"c2f544e2c66f5674e8518845a445ba9bf92fc14256aff00bb2483317aa005ce7","loss":3.548122137708731,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"962c6cc0d3537df019a730bd0500f795c6341e589ca679b6b9597ddb73e08263","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"2eac71c591679e0f23912883d9824cebbda4f055962b173ce1d650d13168c1ed","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"b627705be375f4bc18ac0670ba4b433503ceff0fcb998f0a0632f1a69ccfb859","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"b029a8368d504608419016c926a74d9f771e2e8aacaa0f3443acacc3afed555a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"b8e034566c3e1e2dc5fa2defa3e4d05f8f27cce2c93f72b1e281252e4f2169b8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"5a3f543b2e53490ee7ef6ead28791a2adf42f4a189503f845cf802ca6c295878","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"bd8837f9540872dd939dca69302a81a989f4ca224f5f0facfe0a736aa30fa059","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"b029a8368d504608419016c926a74d9f771e2e8aacaa0f3443acacc3afed555a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"b8e034566c3e1e2dc5fa2defa3e4d05f8f27cce2c93f72b1e281252e4f2169b8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"5a3f543b2e53490ee7ef6ead28791a2adf42f4a189503f845cf802ca6c295878","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"bd8837f9540872dd939dca69302a81a989f4ca224f5f0facfe0a736aa30fa059","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-143.pt","sha256":"d88cd7eeb053004e20e7efb4bc6e5363806b9683a542289229660227bd67593c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","sha256":"274b635002eee2db6434902b8dde5af572ee2c20a94b61a37919721e09e4d1c8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","sha256":"f3ffa8825dbca2d32bcf14e7873b7428bbf559243e8ff16dc075354d926724ba","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","sha256":"773135a800661f5c9f2cf1a8a9d6801d2af84223c1bf71a59bb70ef7c8ccceb0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","sha256":"8ac0892a4701e15dea8de37918fce3ad43e04c88d75f360cabcc5f63e24db35c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","sha256":"fa8b6934c7aa2c177b0c07dd3892457a1e68dd5692c7fb229dcd82210d4de53b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","sha256":"77834d26e766dd87b28160ff8b79128ec70c47aa19266da482c2c935c35e9b89","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","sha256":"a91970b40f38c97e8aec0f545f0b8bcde95c8ff477f87ef6014197cee7270517","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","sha256":"879ac900177d5825f2229ce0bd9d49956b0f8b911d1405eacc34df18b441d48b","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"0043dff9cb4b339f9cd1931508abafd16eae050138376757ab6ff3a1bad92e0a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"ac7903ae2aa6728d3f8a1f43fcec3a882417070bad5401677a529d958f4fd621","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"fb3ba1b3a88b67de804b0c5acfb14d686e8694ba7d88b4c56a290f603938c1b8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"d179cfe306630bd8659f3719913e57bfbe611007a9154abc223c222c4b0971fc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"17c4ab9579fdf615629b8d8cb7f562a22c0ed7d8f191f6c6c5242d3319fecd33","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"b0ace57589c17730d5bf2b46d23bc74c75995ee3d20d61c90bbb9079bec59b6b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"9a74a4fced0377185ed9404a690876907446a19a9411d025f73b9e39692e3a11","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"61e144d90824a1cc2f2369267b3628fdcd819e3415741c7059d52cccc414f05b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"956f7dc0e15107323edc2beb73522ea4312d9868ad2cf467af8223debaf9370a","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"35a5b05e2b5ffa961be5d8e024acf5bab1eeb1e80a7eff16208131cdfa51c77f","bytes":5315}},"requestSha256":"7cd70066d23b33deef970d7e42aa2c95c272b147f473c5d194bf1d588931f50c","metricsSha256":"11d3636af7894919ad12b7fa9f263249d8e15dedc04dd005a36d998f14e1f7ac","continuationSha256":"be777040ce1494fbb629eb5e8bb48e2aad180f57b8291c42befb61072212ce27","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"0043dff9cb4b339f9cd1931508abafd16eae050138376757ab6ff3a1bad92e0a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"ac7903ae2aa6728d3f8a1f43fcec3a882417070bad5401677a529d958f4fd621","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"fb3ba1b3a88b67de804b0c5acfb14d686e8694ba7d88b4c56a290f603938c1b8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"d179cfe306630bd8659f3719913e57bfbe611007a9154abc223c222c4b0971fc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"17c4ab9579fdf615629b8d8cb7f562a22c0ed7d8f191f6c6c5242d3319fecd33","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"b0ace57589c17730d5bf2b46d23bc74c75995ee3d20d61c90bbb9079bec59b6b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"9a74a4fced0377185ed9404a690876907446a19a9411d025f73b9e39692e3a11","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"61e144d90824a1cc2f2369267b3628fdcd819e3415741c7059d52cccc414f05b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"956f7dc0e15107323edc2beb73522ea4312d9868ad2cf467af8223debaf9370a","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","sha256":"7ff53eb2aaada087364b0774ffed537a6c0308f358470e2d16e24903f798a123","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a320bd6bc38c54e847c3074dcfee6a7b5c49d978afb5ce7eb90ac0574d9dd497","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c2f544e2c66f5674e8518845a445ba9bf92fc14256aff00bb2483317aa005ce7","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s43","horizon":80,"attempt":1,"job_id":"2481775","recipe_identity":"bface5c1fcff393cc96ea5b170448e3d0e81ba45d980dcc93063fb3dd46d2880","loss":3.5535805514107945,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"a75e0c9fd354734af6a264397f34958a3c18330af0beb8fa4dfc3d4e62d92e9c","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"ec90bf79cf80d87cda5a19ec85f6371b72c7a96b8c0516bf7eed41534d55d0ba","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"84f7b23bad4a440a8acfd98d422470263a4d5580578f8f2c54f404918e1e99e4","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"315fc1783f7aa01654257906b6cf3c6619b13cac1d373e9149931173cc13de04","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"a81cc8449919a3545cb88cf4c3cecedbbb6c4c769e8010830ef36d1c0d620879","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"9038cbbff5b2944f7bd4aeef9aa262bc24f8c03f379b4e4fbc41754d6b701b35","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"c17ac0e2071371a6c964cb7706f3255d618e86bfca9f55b874e63d08b8853f01","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"315fc1783f7aa01654257906b6cf3c6619b13cac1d373e9149931173cc13de04","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"a81cc8449919a3545cb88cf4c3cecedbbb6c4c769e8010830ef36d1c0d620879","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"9038cbbff5b2944f7bd4aeef9aa262bc24f8c03f379b4e4fbc41754d6b701b35","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"c17ac0e2071371a6c964cb7706f3255d618e86bfca9f55b874e63d08b8853f01","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-143.pt","sha256":"3aaabe9b20e47c7e5e4fa3abeae68be62d6c8665f734f5eb732c915acebf728b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","sha256":"de47867ab53c382fad93d34f2785e146bf75403bada233171a5ef0a70be64036","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","sha256":"8b0078603d1eccd8a4ab659005b6c4c78ccd11c6d38606cf7d5a07be045acaba","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","sha256":"d86e0febcf4c41a4f53926129cc2837d7f910af9c6ae1b8d348e49661160d304","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","sha256":"ebbc61c1238a4efe8dcf4e73f515ae477144345bde4f3a65bdc2571526d22137","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","sha256":"f83d937ee9f0fc3fa1b435a6fb15971cceeb4671a2eb537dcfecd6485120f840","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","sha256":"d32e0ad04ba6048534b1b424e90fbd18bd13ca8b3b678851484ff09918ecb632","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","sha256":"529adea3a18ddef1705d0f4fce4676485301125f70ab346971d5a8ea1a1c232d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","sha256":"e713f20fa609030d8d27970632e2d32640c86b680d2f10bbd2c1dfb140aeabab","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"3b89ba470428dac7d48cd156e77c14d8fe2cdb2ce372098631a7ac51fdf58fec","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"772724a421fdbe4eabad5fd392b7f9a3e178d5bdeaa1686188bd7a0e90ebc1d8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"ba32d5d341445c6e49e91da99f7d9107fa232dab4b8d39fe0145875e2474f847","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"88487699f2305748d17e2635e14043db32efdf44fac8e2c1b068a93e8d9ef3ca","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"5111ced37031b55c4a099724577dc729ddf2e13cc1d09dd99ae50f79e127ef25","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"7968d740e8382474538e95f35246b36fc63bdce6874be4274c3b072ee853989e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"1683338eeb1d01c54ed10651024f4d3d65e2806ffd6d1d75aeeb2eca0fd10e1b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"ca0daee44a57f68cfb67781e87d40ede9c0c7a8b660fe76e21986acd5e2fb1f5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"540d16a81656db45f34a9efec7003227c6cbaff8104d36bf056d6f636d453b89","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"bcfbc387efe9b2fde077c7068d2d9fda220cecfdca7227d97a4d6ab09b50bb02","bytes":5256}},"requestSha256":"633192863c0b8c7757ae80bc6129516d32af3900f94f0eefef632021ecdec4f1","metricsSha256":"bb8887e7fccc686e9209fbe8d7cd2731fedea9989a9f81b31688d64e34ca8e68","continuationSha256":"127e186e2a0d895d7d6f0f32cc6b1a59fddd5d86722ad4d93576f0ef89932492","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"3b89ba470428dac7d48cd156e77c14d8fe2cdb2ce372098631a7ac51fdf58fec","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"772724a421fdbe4eabad5fd392b7f9a3e178d5bdeaa1686188bd7a0e90ebc1d8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"ba32d5d341445c6e49e91da99f7d9107fa232dab4b8d39fe0145875e2474f847","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"88487699f2305748d17e2635e14043db32efdf44fac8e2c1b068a93e8d9ef3ca","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"5111ced37031b55c4a099724577dc729ddf2e13cc1d09dd99ae50f79e127ef25","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"7968d740e8382474538e95f35246b36fc63bdce6874be4274c3b072ee853989e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"1683338eeb1d01c54ed10651024f4d3d65e2806ffd6d1d75aeeb2eca0fd10e1b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"ca0daee44a57f68cfb67781e87d40ede9c0c7a8b660fe76e21986acd5e2fb1f5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"540d16a81656db45f34a9efec7003227c6cbaff8104d36bf056d6f636d453b89","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","sha256":"ba14060a66456e611ea6d070800984b30da27f9e4b30bda0905aec48d35bb839","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"226e88acfb25e532f321fb3f5aa910cf80f1f0f25cbcf82ee3f0f281ded3a2ba","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bface5c1fcff393cc96ea5b170448e3d0e81ba45d980dcc93063fb3dd46d2880","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.006-s44","horizon":80,"attempt":1,"job_id":"2481776","recipe_identity":"14c7d5939ce93fbcee6c6da7de29a9b2d332d67a1f42e78fcf32b592c2aa8dab","loss":3.557084646358441,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"74e5cc2666904f6aa643313d2212002b2224e3265a0570dd37a98656ebfc705f","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"820dbebd90b8ff61d5f1844bed68951b8586b34622ee9a6ae72ccea6011eb9b8","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"4ca85180d00053df24273c5f14034b7c7a817e73a6f67c677929fad44594dd93","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"ba661375d5124ae81ea206d078df9c63a4a1916d542db2f4669a2b74a86af882","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"ffa0d1c8da75cea6cfbf18ad37a8d2947d98072193fcc1b15355e4dabe518765","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"28370ef9b2dbb25e8caa60b00d7c000bd659fbebc7cfa23255fb47a72dd0d07d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"5c2e271f501a12b4568c58fec748055dee1ae6915edcbd2eaa45bc53bf9482bb","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"ba661375d5124ae81ea206d078df9c63a4a1916d542db2f4669a2b74a86af882","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"ffa0d1c8da75cea6cfbf18ad37a8d2947d98072193fcc1b15355e4dabe518765","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"28370ef9b2dbb25e8caa60b00d7c000bd659fbebc7cfa23255fb47a72dd0d07d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"5c2e271f501a12b4568c58fec748055dee1ae6915edcbd2eaa45bc53bf9482bb","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-143.pt","sha256":"4f621511c317bb23ead00caaa5a97e0edd2932821e13fa3d58000c851172bcf0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","sha256":"259edb1266be2b5eaf55ecd1f0009f989c99582cafe255e4cb870ec13950af41","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","sha256":"9fcfa5e38f80e512b0f07e5a3140ab224b306581d97ad219652462763c2c0517","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","sha256":"b593f56ef87a2b67cb220525d7b261f30f028fde2cd3371c9dae5b82140b1e38","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","sha256":"d67bd4cad5ab16bd283db28123f99b71397fb49d8b3ba3f11cfc5dea03372cee","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","sha256":"25fd2d8fb7fd178bbe691986cd48f96ecfcf75f10bfc3ddd80d44b313e254cd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","sha256":"094550b99f88bde52e6cf910f6cb06fa6ab3e3e1820313f40885cd7462f0310e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","sha256":"9ef2f737dea39207b8086238118a9a189320186fa01dee7e0f6ef57b2cb8fc8f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","sha256":"64bae382b34ad2067cc62ab2219ebf29d194bfacf7175c2fab277087e72e19c4","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"e9d639fd44e11153a446aafe78eb33ed443e775069037cd4684294717314ace7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"683527d26c2050047416e2da913f24cef24b94348907cea603e1c7da229801f4","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"7f48324975ad1ffdf6b07c942f81b38bbded38c481cabbd38efc1f1ee42f7766","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"02a5a676a11b0f38a5b80fd3007d6d183452a72dcd65c311e55b91ed744846db","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"80528f822202c63b7b42d66a3753b17691569b022d34c22965320dd91bca6f4e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"4bf4c04a134b1c89f6eead661a666cedf08788cc57ceb21f7490372bcd97195e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"b994b9841755f63409631365bd97433329c15495c24b8ae00ec736f2a4b0b222","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"99831a08373b7710493ff7b63d3a58552cabf47d0a562393270cb4d96a89d024","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"728c3ed4776f7cdfb99d9e10ee908cb01b97a64eb5abad626b8e68ab24f01a1c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"f0793d26f91bebeb29eb01ddac8470cb91642720c980f10fd6d6d2f2424eeb73","bytes":5313}},"requestSha256":"0f04033dd39ff477ae3ffd40fabcc239dfb132f009df3871899867fc1c9c7916","metricsSha256":"20a84512b4111fd3d40532f1eeba66b99aa40a746b04f44a1f904c197ef20bff","continuationSha256":"173f2745e34f0375c2abf90a04c52a8ebde425636633bd579e04ec19203a0c51","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"e9d639fd44e11153a446aafe78eb33ed443e775069037cd4684294717314ace7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"683527d26c2050047416e2da913f24cef24b94348907cea603e1c7da229801f4","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"7f48324975ad1ffdf6b07c942f81b38bbded38c481cabbd38efc1f1ee42f7766","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"02a5a676a11b0f38a5b80fd3007d6d183452a72dcd65c311e55b91ed744846db","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"80528f822202c63b7b42d66a3753b17691569b022d34c22965320dd91bca6f4e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"4bf4c04a134b1c89f6eead661a666cedf08788cc57ceb21f7490372bcd97195e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"b994b9841755f63409631365bd97433329c15495c24b8ae00ec736f2a4b0b222","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"99831a08373b7710493ff7b63d3a58552cabf47d0a562393270cb4d96a89d024","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"728c3ed4776f7cdfb99d9e10ee908cb01b97a64eb5abad626b8e68ab24f01a1c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","sha256":"30b57f27250149a4b907027832deef4d10bd547ac1e5b31c1e40fae4fdf1f288","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"14cfe52532738e879450ff01defeb4d491386b29c1589e5d19e5896da721a759","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"14c7d5939ce93fbcee6c6da7de29a9b2d332d67a1f42e78fcf32b592c2aa8dab","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2481777","recipe_identity":"8b9d3e6341754ac3fbb7e37a4de8ceef881d9f454c4db6fd9605a4d8161daada","loss":3.571304422519585,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"b957d300fad72d99664489e4cb0585bdff321688a797d5ec25bc91c8e28cf6d0","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"2e00b23b192b6130825dd065084a190df19bdb1b1c6ad7130804b42b4c1f4d81","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"144f2af75609fad3fe0b01e39220655cbe4f93477b96d332c7d950cb9e5a7d27","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"2f39831457ab29103bedcdf375682b4a934eec3fd1ff660b1ac9264ee1e5d11c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"a6854f06f1ce4b7c46a1a388902313f77d911ef110d134a964391d32160e04ea","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"3ec3385b3d9d7c9e4f8ff19c870fe2dbd8a56a7f93d167d7dcca36ff6f4890d3","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"78138655bec3fecc6862b8230699fd12d038917f8bb5378295c93a757319c994","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"2f39831457ab29103bedcdf375682b4a934eec3fd1ff660b1ac9264ee1e5d11c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"a6854f06f1ce4b7c46a1a388902313f77d911ef110d134a964391d32160e04ea","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"3ec3385b3d9d7c9e4f8ff19c870fe2dbd8a56a7f93d167d7dcca36ff6f4890d3","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"78138655bec3fecc6862b8230699fd12d038917f8bb5378295c93a757319c994","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"3276cc0ff342290278da275d9bbb3304ed80a2cda6a63b3d2f894e9c2fe3ef45","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","sha256":"3862257cf8aed0d0b2dfe53b06874a9fadfa144175dd96275c6f8b56523a3bb9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","sha256":"df46359a3fa9072805d050e62ceef9113480def0072bba733271fd57bbc2a90e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","sha256":"78fe529e3fb95a5415e5f2531dc1c6608edf0c189863211e82e1fb359e588651","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","sha256":"6dfe823d8af160e6a11436e7650858053ce59c9246987c1e95a752e5c3412fae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","sha256":"b7ed9d7bec59cfdbd4d7459a5fb10e1692e35f0bfe501b31b6a818387608009c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","sha256":"8fe249c6c2c78d6c18c53d885d1284e9210c67c3983364b824613372cea5482e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","sha256":"9f7ec722f91c65b001be7e6502dd5a3ebdc9a3be9ca0472b1848dc409f79c13d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","sha256":"6a5215ea29902581b8b8ca32072503f701d210ff0b1e8352e397eea39fdbb9cd","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"b6bec49b13d8ce9425a129c49711c09e4daf6d4213f9f1be0b03b738aca3ff34","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"68ba61ec9cff8355cc4d7061ca41961fef8055036d9ef20710b84cf8649a8d92","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"05df48b7bd6ec13e7641c167c994bf976b409d7dfcf4c8272de538f70a63171b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"96aca5e3586052d16ac3b8e97c9fbe010c5d833a1baaa7962bbdff4dbb3248ef","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"6d57af0ca127dfb6162457fbcaab3bd6601ef778e4a6914acd6b49800c7f9920","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"ab19aa1d6210e10c0ef67e58387ee31d37c165f15f38c936dcb49ea4794656eb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"030062b83b9f1d34c8a1fff22ea7d28f5200fc24069bd4c84e96592b8ed38b13","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"98fe7c4afeba872a76f247384363d20b050bf39fb0f36d14b0fa3c7c89d39e07","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"511270b9f23d1c207ec2b7d1243af01c9d942f5d6c55332f8aa4c24fdfb9df14","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"a1fc88e216705700ed3a0bdbf831a54cf8c3bd3020d26df6cb404f3e0ccfb46b","bytes":5256}},"requestSha256":"7096fbb8ca1f8da2831eccf1c2f5c10b1fada37a71ab5cfcdc4085e3439cb624","metricsSha256":"1933742080565d3f8a927d1201d185aab54362111c77b677a36f83596e722c56","continuationSha256":"67e5ab29e7e452e58678caad7d7560aadcc9500baadcdc1419332954502680fd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"b6bec49b13d8ce9425a129c49711c09e4daf6d4213f9f1be0b03b738aca3ff34","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"68ba61ec9cff8355cc4d7061ca41961fef8055036d9ef20710b84cf8649a8d92","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"05df48b7bd6ec13e7641c167c994bf976b409d7dfcf4c8272de538f70a63171b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"96aca5e3586052d16ac3b8e97c9fbe010c5d833a1baaa7962bbdff4dbb3248ef","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"6d57af0ca127dfb6162457fbcaab3bd6601ef778e4a6914acd6b49800c7f9920","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"ab19aa1d6210e10c0ef67e58387ee31d37c165f15f38c936dcb49ea4794656eb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"030062b83b9f1d34c8a1fff22ea7d28f5200fc24069bd4c84e96592b8ed38b13","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"98fe7c4afeba872a76f247384363d20b050bf39fb0f36d14b0fa3c7c89d39e07","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"511270b9f23d1c207ec2b7d1243af01c9d942f5d6c55332f8aa4c24fdfb9df14","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"83aa390756513077deb4465b95cb79b4a4144f9517fd176af70c83e35ceead6b","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6c946f13bdb7d58068672c95356b77c1e0a2a13a8eedd488308a7a04ea6a9a6c","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8b9d3e6341754ac3fbb7e37a4de8ceef881d9f454c4db6fd9605a4d8161daada","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2481778","recipe_identity":"fb7b4e1d5b21826480a1b1f827bf340bb8a244d2e14acab2287526502d404cab","loss":3.5640851911404656,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"ae6bca1d3c644cd5c31e116c1e818bd01c8c82194b2cafcb547c0cfa3cdafa2e","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"c3165e04cd80b856e5688eb0cd00e2a2cd5e45c8f8e6ddbe9b78be80677b446e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"439270dcecad91cd3a4074d9773400b95607b8fb09e1197112b62da8738be834","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"433edc3c6c03013c7a9c779d8024b36a8e1d0811d690e1617ef1f82418c1ea52","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"0f66f70413fc6af53f561110f55419a59cca0890043ac00939a934c77cce8558","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"488a2ec1b617bbb7e98b1024cbabbe6ba464eaecf99ec9a1dd7a6aa092b782a2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"fe7c71ceef6e2e554888c2a4fc90d296c50a291feaf309a2d20098a54eb39775","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"433edc3c6c03013c7a9c779d8024b36a8e1d0811d690e1617ef1f82418c1ea52","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"0f66f70413fc6af53f561110f55419a59cca0890043ac00939a934c77cce8558","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"488a2ec1b617bbb7e98b1024cbabbe6ba464eaecf99ec9a1dd7a6aa092b782a2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"fe7c71ceef6e2e554888c2a4fc90d296c50a291feaf309a2d20098a54eb39775","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"8888706ae92296be952c16ba899df598913c448440a97034828ae4cf9d887441","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","sha256":"9cede97f7f57e2c6f5a88d61d7c48b8fb07c75d504f9ad9688b0f00c598bb262","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","sha256":"c304922f5f6af1072e86da9efd408e61dee92d47f20365f931276a98227d450e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","sha256":"97eb1e7fb1f3da52d4ea13c8c79b18609a48f8ee0fae00b0ae40c042f0a0875b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","sha256":"e45373a9fba76821db4b0d7be15cbae3dee29e977cd99cbff8bd7c72f95a30fa","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","sha256":"e768ebc225c1bc81bda8a936ff44c28d804c6cff459f144c6fbbb2aa6f65aac3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","sha256":"81faa08da92d361426f14df21a214e7c971755a14dc7a9f276066bc245d11240","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","sha256":"c4f02b62768f1c90abe36f9fa01a0a56106950f982fbfe8eaa6e911d8958561d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","sha256":"2bce235bdaaa162293f559e46a9d0bbcf95f209a2d37c98e4648dd08a6ae59ab","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"c128f03465ea7c9644f5ae3b1f748e1945fde778001b2c0406f6b8f468d410d0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"36ebdb6c4a7339a30ee5323d96dd0d5e18d76ea4f567dbf8ad65a5f319078467","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab8364d880fa109782c8a6d4327d03c106df4dd695d04d03aa40bc903f4461fb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"dcb248a6167001e3aa1fb4cbaaa680a64a578ba3ea5eb5e68edc1fc4bac6a9e6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"d4f523f44c5df3bfd017043c216719a368c0b06ab1f9d6982860fa802d2f19a3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"e005bff2f8ccc052bd26992ae2d6644f5a774737f24d6b80d83e1a7f4a260a65","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"269575e26d8da1cf5c63cd2488734521d7f463de07f1fe7aaf156bcce24797d7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"bb4add6681633c178fd3348437d7e386a54c9ff7f695cfa4246d13fcd11ce0fc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"e7a51ed8930518698a489ab609b46bc77d8d551f016b70f0f2efc4805e9b379f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"06255ebefcb06954ca1e59a4a9da0f41b2c18397a6708ce54b059191ce78a1f2","bytes":5312}},"requestSha256":"a3dbf2e67b6c7bf860e52a02787ba34cfdc25d960a09999fb4f276be56ce9df3","metricsSha256":"0f314808abf576a4d11681b7ddc44fbfb9cbdd063264435c3176518c0941aca4","continuationSha256":"c4bdbcdf161f3ebaf09552ab0b2b5276e6cc6ae1316299eac81ee9bee8586364","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"c128f03465ea7c9644f5ae3b1f748e1945fde778001b2c0406f6b8f468d410d0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"36ebdb6c4a7339a30ee5323d96dd0d5e18d76ea4f567dbf8ad65a5f319078467","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab8364d880fa109782c8a6d4327d03c106df4dd695d04d03aa40bc903f4461fb","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"dcb248a6167001e3aa1fb4cbaaa680a64a578ba3ea5eb5e68edc1fc4bac6a9e6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"d4f523f44c5df3bfd017043c216719a368c0b06ab1f9d6982860fa802d2f19a3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"e005bff2f8ccc052bd26992ae2d6644f5a774737f24d6b80d83e1a7f4a260a65","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"269575e26d8da1cf5c63cd2488734521d7f463de07f1fe7aaf156bcce24797d7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"bb4add6681633c178fd3348437d7e386a54c9ff7f695cfa4246d13fcd11ce0fc","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"e7a51ed8930518698a489ab609b46bc77d8d551f016b70f0f2efc4805e9b379f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"2d0fab72fb1ae763f59ceeaaf30ff5387ba5fb40f5fc646fc237f522c5420e8a","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"863ce256468e42d5d33283d28a961a886f4f12f77c122368c4c24e767fbf28f6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fb7b4e1d5b21826480a1b1f827bf340bb8a244d2e14acab2287526502d404cab","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2481779","recipe_identity":"c8a2d2bf04e6632cab7028709ab302f9cec34a8ed8441dd751ba96f9d0f091a9","loss":3.562859040919509,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"086ca3b7bf07d4bbde2378e8dc7ca825621df206e1d6318144da655dfeb14b08","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"a0fd11d1d26b8c590f296f11621039683c2a997a28bf5efc99526765387ce458","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"f9abef672153c12676f2cdc87cbc2d18a98769ab5c80d3edc09ad1781269c1e1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"b692085f63cebd311faa4d0d7c72c3f67c4f778eeebe24ef2bccf6ba7e372afe","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"2f689a020e8de4402b276dd855467666a7ad9f8b653f8e0ed67ddd3923434d91","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"e0f7c7a7773efb1345b9f712fb4a2dbc8496c43df29059b512ca77db078592d4","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"d8f502106e7a73223fa56d132175351bf622224bdbb6a934e02527d9000cd67f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"b692085f63cebd311faa4d0d7c72c3f67c4f778eeebe24ef2bccf6ba7e372afe","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"2f689a020e8de4402b276dd855467666a7ad9f8b653f8e0ed67ddd3923434d91","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"e0f7c7a7773efb1345b9f712fb4a2dbc8496c43df29059b512ca77db078592d4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"d8f502106e7a73223fa56d132175351bf622224bdbb6a934e02527d9000cd67f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"b127ea8c7b91afc6bb7a78bdb0ac0ac28a029f16e2b4d5030a0cebd6786b2f47","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","sha256":"963a996b4937c334810a5248c0995b8245dfe28fdadfd1bf77bb4c0512aebae0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","sha256":"b38dee9752f76da30f3e31d28ef18385fd4a1d3558962764703022076986a217","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","sha256":"810eca06aac8daaf7a3704a2638457264234dc69f95e9dd71b807364162a0d1a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","sha256":"4c3c15233789e2a5f7ff80dc9f4fb64a81b002ed2f8e7ca3c5b0dffeb471d82b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","sha256":"17035b17c0ea49d4234a858bf37298e80d5139f300cc2258c7518175afa0963f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","sha256":"b7d2a244ee752ca3cd13f4f27a524646028803be23b31f740fe9e9e23fae59ac","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","sha256":"1152a56b075efe735e8b12f8d79c9470de24cee58a868fa53b32457302acc13d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","sha256":"93d72934e799a7133b484a7ed3ecb7b7f66858f3da8219131f2cc5159e9ee76a","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"49b7cb5e0d0ea908fc574223c7311382f46c9c2f669ae53a7e78863242bbd555","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"3ae44499815b8fd4b8c5f86397c5fc5dff61f0c7729c051dd463da72a2119790","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"250b26dd56fe2352b11b11ca6554a8bfba39db9802b4b27025162503f75eaf83","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"e0f895ba7c957c5272f30aa547d4d5cbf83e75a9f081bac01318e209cf959d87","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"15ddf3987e3f8f1482b6d0785612acaea275c95bc653e1bffd3c1a709f9ece2a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"9e24fddacadd8b1cadbb97f89947eb53f249c68c2c3b50cb72908dfc35febbcd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"b1a35b29a1f14d507fc4bfbc5d672a4c1c8714eac3ff8be8f1a5a57041eab6e9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"a5d2be4a9159ff1d2f5e77fa70d337296db6f8db5b52fb1c17fd18798ee701bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"10762a8cd2e6e8b8ba25df004b091e4b134862628d810821c38113953285db72","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"617d41d0af358ae4c125f6055630a930335d557b3ed0dfcace0077591bf3c7b1","bytes":5342}},"requestSha256":"24097fc9a662bb905eccdc80da1d32785c79de64f2193f51a6fc03fcb1d5fe96","metricsSha256":"b557355f0a5f3f2149929a49f9abd9676695cf4896aec7e5295c7764a638d8c9","continuationSha256":"7a43f62d018ae1717ce46c9efe579340aacc53efb40248e789d78afb89e32900","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"49b7cb5e0d0ea908fc574223c7311382f46c9c2f669ae53a7e78863242bbd555","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"3ae44499815b8fd4b8c5f86397c5fc5dff61f0c7729c051dd463da72a2119790","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"250b26dd56fe2352b11b11ca6554a8bfba39db9802b4b27025162503f75eaf83","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"e0f895ba7c957c5272f30aa547d4d5cbf83e75a9f081bac01318e209cf959d87","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"15ddf3987e3f8f1482b6d0785612acaea275c95bc653e1bffd3c1a709f9ece2a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"9e24fddacadd8b1cadbb97f89947eb53f249c68c2c3b50cb72908dfc35febbcd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"b1a35b29a1f14d507fc4bfbc5d672a4c1c8714eac3ff8be8f1a5a57041eab6e9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"a5d2be4a9159ff1d2f5e77fa70d337296db6f8db5b52fb1c17fd18798ee701bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"10762a8cd2e6e8b8ba25df004b091e4b134862628d810821c38113953285db72","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"d6779561b43bc73eb7348feebf5c0084790f76d576909963efb27593c5b626c8","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"05ffdc6f00a8cf2b69a2cccf11ba828b2a92b0a144b707559359e67e3b519660","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c8a2d2bf04e6632cab7028709ab302f9cec34a8ed8441dd751ba96f9d0f091a9","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2481780","recipe_identity":"b4a8835d20266caa506a44d44d36ebc8ec5aec2a9c926f371458e0fea045fff6","loss":3.5737734380320734,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"52b6d6904a179c3d3f2ed3887012566fcb06c391034f579a08630895344ebc32","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"ef2cf442686572dd566752dfaf2f1c5c0f119ae37cdafa48065fbd29e0e59ac6","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"ab10efeb290cd1941287a89bbf60f0936c31a3562a91ae1bd0a1e757688e0ba1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"37cc6bfc438713c0f03df63e785e6f62b450e21e6676b7aa9e1f652c33f80872","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"444c3b4bea28db7f872d047e7c2c0196d0a2dd14e0b8b5103f22049ebd87c435","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"c822e1c1cea847678b7afe14d0499219bc3e68ad73d845ec56746ca9a4f37da2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"e823b1aac2347994e58b014754dd0d6baba38284aa93f48cae08a0b6000dc588","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"37cc6bfc438713c0f03df63e785e6f62b450e21e6676b7aa9e1f652c33f80872","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"444c3b4bea28db7f872d047e7c2c0196d0a2dd14e0b8b5103f22049ebd87c435","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"c822e1c1cea847678b7afe14d0499219bc3e68ad73d845ec56746ca9a4f37da2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"e823b1aac2347994e58b014754dd0d6baba38284aa93f48cae08a0b6000dc588","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"5ec8c86ad48c270bf98f0f956beed43a55b11dd7288e46e731705098688dc2af","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"204d67cffedeacddb3e350556b05677720ccb39fbd61469113089aebac104d71","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"eddaa33c81490b3e551bac6bea1c829a4dbff049dec885875294f01f81b42245","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"8f11daacf2f76b7fada67501929d5c3931f5241bf481acbc4da72b2a74bc80d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"4fa0b8842dc0f87bd6d063702c0673a464e6a5ee0380d6a38d5e0b8ce8c0981c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"bd1b5f22119e8e35beffd600e025791bdc8ef0b86840e306217475ac57d339f8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"3c6f684c409a892d7dcc64f17b16f80451bd162e651e95517e4d923586c1b5e5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"49a9ba0e7d395c870ac71db6cad403fc869d18a57e2d04a09130cdbae0ce536d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"e6d81fdf26879eb23c26377823aed1ed29abfa177753351fa130e00806c3a67e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"ae80db95877f6fe838c19fc03d3976d4539c13e8af2ce080c9c38d3f267a95f3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"82e8a7ea2eb2bfb7f86c5eb8c354862ff87482b4d36c9023b6cf60cfb83c5a79","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"8e61226639d8426d0efd551ee705dea96e907d7e5f0dc4b1b6c3961baea501a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"81bac4bdda6f2c20ea645479b489aa5dfb41196097e50c1c8b667b8dae4fc5dd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"60ac469ef285ecdfe05ff36e7757b5e03cca76e380bd0e24c4ff0e58c3810b36","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a8ffd806c163dde69e7e063b7975185e72e46c86ca84924fb876870e858645e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"4f194addf19d93da0ab8f838da3f3bcb94fb597f7c522ceb36a4cdd09ba1358a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"cf777c8863e2c6357a704c2154cf0ad7f3b9e092d1b79019b1d6d62263576685","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"0634a5608618ac8c911ecc1d654368bd29d9df849917c51dd30f8dedc4b41d08","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"508b10c62f555fc8da6db69983ffb03f8955aa62018f6109c55ae44ff6a35398","bytes":5258}},"requestSha256":"f314ae0b374edb02da505693933d0a6ce2c46f151f360fddf4dd5df74746503a","metricsSha256":"062d2adcde003b00fdb54925392b2201e65c580db5839510ebaa9c7c6404dae2","continuationSha256":"b1ef7c20e7cf282466d6a4babeefbd853db39133e2f22855ef40a17e603f8b6f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"ae80db95877f6fe838c19fc03d3976d4539c13e8af2ce080c9c38d3f267a95f3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"82e8a7ea2eb2bfb7f86c5eb8c354862ff87482b4d36c9023b6cf60cfb83c5a79","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"8e61226639d8426d0efd551ee705dea96e907d7e5f0dc4b1b6c3961baea501a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"81bac4bdda6f2c20ea645479b489aa5dfb41196097e50c1c8b667b8dae4fc5dd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"60ac469ef285ecdfe05ff36e7757b5e03cca76e380bd0e24c4ff0e58c3810b36","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a8ffd806c163dde69e7e063b7975185e72e46c86ca84924fb876870e858645e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"4f194addf19d93da0ab8f838da3f3bcb94fb597f7c522ceb36a4cdd09ba1358a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"cf777c8863e2c6357a704c2154cf0ad7f3b9e092d1b79019b1d6d62263576685","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"0634a5608618ac8c911ecc1d654368bd29d9df849917c51dd30f8dedc4b41d08","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"f8b1553b2988205f3d96c185a81aaef70a09a39561a9d48eca05342d940878ec","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a9b11b7ed0ea6407a5d7e5028402cbea7c01167fd5a2b9f86ed23da930774f3f","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b4a8835d20266caa506a44d44d36ebc8ec5aec2a9c926f371458e0fea045fff6","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2481781","recipe_identity":"d78484fbdbbfed0fce0a944a30c0725e37d82d083d81e98013c0f3e6b1754891","loss":3.5600984304816894,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"c9f28fe5ba763384c96c30ca35cd8b5e7497c9365e761c69427345a0a13332f0","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"a22b2e29636c1dd00b707d4328906de481adbc2fb8c4355d01e65d1a9c5eac59","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"b8c379015a94e0a1e2e413057e906a83801b43a97474f8b4e3c8961bb6dd7852","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"4e990052cca91c024423d2f60a597ca488eb93e05e619026e54728669a77ae22","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"c181d365b6d9c303e2b2c7572b44448ce4c91649d2c6174874def0e46bf766da","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"f4cc8a5f23e80890804151303ddff9f942bc292aa8dfeb5d797847fe3310799b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"48475c613d33a867646c89672e81ebcf844a58fbf0dea6237d7c71143b7ef268","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"4e990052cca91c024423d2f60a597ca488eb93e05e619026e54728669a77ae22","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"c181d365b6d9c303e2b2c7572b44448ce4c91649d2c6174874def0e46bf766da","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"f4cc8a5f23e80890804151303ddff9f942bc292aa8dfeb5d797847fe3310799b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"48475c613d33a867646c89672e81ebcf844a58fbf0dea6237d7c71143b7ef268","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1e7e011fed240fe4dd13832ff69555f42cc4a107b6f0acd559780c9b35821ad9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"8938709b5c287a565281513547099c97a0655e8f423a95ac2b0084a4cf9d2247","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"f377dfe8f5808febaf668ec0157591765b3600a1e1e5bbf3a8ef65c98718c4e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"ff2b8c4708ba1c92f53dc3ed1f70a7109d123dc1478d0fa4359aa85cb3b321a3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"2732732e7062590b547f10877b5c8b306008dcc70c3f4a621b324ad165cb9ca8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"5f26a0d65ea4929af329f8e22fafe5067f2aa6ac82e74e493b8a2c4db2351e3c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"f281b533cdfc9aa1836650ab687e0ade19f5f00260eab97491c4e82c417bd7dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"16e58a8da61f2ad686bcbe3822ba58f0be02027f65623636d7bd4bd089c5b2b8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"ab29c45c729a655b53b46c7de1cdd280c46bb616298ff5e30cb011196e80e9e2","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e86310c1d17c22b923975e3878b9cdbd4a28e30fa3909391d163405583eb0893","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"0f6f8bd8d960532ec7f9dab909efd5ea023ff234db498f5f2a6e4c3af3f44c28","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"eed2996a5bf3c0fbba9a651d4861bc3186e5890d4d9ca3758882d5b7dd52cec8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"c8fa62e8110655baff68d8ff1cd298f9a3dd6a0a0d9cf05c60cc35863b1adad3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"df665051f11c58512efff6829b1a2a467dadee4f591f081130f87e0ae834ee08","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"f8d9dbc3751b45416d0dcb99780577c97b4ab0ac71d7b53e7d433b4b84f02135","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"a5e5db1afeb1a08129208221910bf5f2899d15682d1c1437c2543197ac7e0099","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"93e915bcb42685968867d4461da3db34dbba9b26f9fe252fa74437f46ef39b79","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"96e74a35139fcf042e5b0a8a4d72ec42210c9fce087b3c6bce6491fd67da9adc","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"1c779a5c7d1dfe26d5d334622642bd9bb034c700986c678cfd8858ac75a7c307","bytes":5344}},"requestSha256":"2cab168ab88520ea6cd570e593d9cf5772f34ad22f6406e73bd75f73677ad7be","metricsSha256":"f31231770780b0e4d8f5d1cce4868925aa08f34f3215143e596cac381228fb24","continuationSha256":"cec605f9a10aeff57ee1ed0e309980a978fb888d644483c9281007011f378a34","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e86310c1d17c22b923975e3878b9cdbd4a28e30fa3909391d163405583eb0893","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"0f6f8bd8d960532ec7f9dab909efd5ea023ff234db498f5f2a6e4c3af3f44c28","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"eed2996a5bf3c0fbba9a651d4861bc3186e5890d4d9ca3758882d5b7dd52cec8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"c8fa62e8110655baff68d8ff1cd298f9a3dd6a0a0d9cf05c60cc35863b1adad3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"df665051f11c58512efff6829b1a2a467dadee4f591f081130f87e0ae834ee08","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"f8d9dbc3751b45416d0dcb99780577c97b4ab0ac71d7b53e7d433b4b84f02135","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"a5e5db1afeb1a08129208221910bf5f2899d15682d1c1437c2543197ac7e0099","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"93e915bcb42685968867d4461da3db34dbba9b26f9fe252fa74437f46ef39b79","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"96e74a35139fcf042e5b0a8a4d72ec42210c9fce087b3c6bce6491fd67da9adc","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"6f6a99375291f618166728a28b33a61721c9d4000a3594388ea6a48a32db8b00","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f70909a0deab7b2a89e2803a689f9f1bbc7f23bd7f1a54f587f05584ca0967d0","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d78484fbdbbfed0fce0a944a30c0725e37d82d083d81e98013c0f3e6b1754891","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2481782","recipe_identity":"726114c13eb4db67a0b450a5336e5e1eb52e3ce013b73a721033f9921c2c9a19","loss":3.5680965025025033,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"1c21322d23f0007daadd74d998fc1d4de0f964b74f31ea9808b6dfbcdf631bf5","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"0eb06bed4f6cd09a2aedbfde8307c94873ef1b1a3cb811636aa7fd6d26c53df2","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"e3ff7e598fcef88cd80389a14923d875578510c6cf47c7be409dfe909f83dbfe","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"85cfd6003968ddb593d0bf71b3c8de7fe4e837ef9221f1ec563649d4aeba9f90","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"179eb788342af78707c040c5d21fd3f9b1371d8d5e5cdea222bae099f5931280","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"3e1309691b2850c3b426c2f28db9dcfb317638aba6dee25271983b0fd9aefd25","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"e1589ce287e8bf4c00be964a8f101b116d5703df2abf15511cb287b17c1ee912","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"85cfd6003968ddb593d0bf71b3c8de7fe4e837ef9221f1ec563649d4aeba9f90","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"179eb788342af78707c040c5d21fd3f9b1371d8d5e5cdea222bae099f5931280","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"3e1309691b2850c3b426c2f28db9dcfb317638aba6dee25271983b0fd9aefd25","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"e1589ce287e8bf4c00be964a8f101b116d5703df2abf15511cb287b17c1ee912","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"9b50335c78a0dc2624e460cb4721925161b5a336ad453f8e95ead676bf48d065","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"d44630f016fdfd1749312d942ac032101087550796f19b16f521b2b086c20913","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"a36eef239b6d41665aa1df58fa87dec418b5560818d54ffd122d48b51da5f735","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"e15c8f4628bfd4f642bd615412a053129bf471b5dca164a176ac70c8a06d36c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"fb4ba1d384eb5cb397b2d8a96f1c9ddc3282d2e955c16a355586485bde433e6b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"c8ec679e8e813e099d5e66b3b4e1a9b3267b13b9853bd224aeb54264e458e525","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"2899c1e45f25147e1493e3c1f920152e830a9a41611fc3e2bb00f8f2453552d3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"e3b097e666b3912ef929dc4b3824a2bd8cf1cc2f88ab687e78d7e7d866347bc4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"febefd6761325214893a6a93f53e39bbe8ad1fb4fc2dcb4f339d3576e4be0357","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"3da15c04f0c4e8ee6457edd54d27d778be0ff11cd13ee60abbb301c16d957679","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"7ab01b699e7912f3b886c22f20e9de01f8fc9e2ed2f72aab998f65d926b57056","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"5c9912b05a38b4629e9e8dc8128d60c59fbab85b4f582e5aec8d95f169f44e65","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"a6d1719ae88b6d1abe9dadf362cadc3695b2351c26ea0056aa9b4654d04728b3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"f00930f538d1fd449b85c3c1bc20af2d0787e37c4749fa175dd00866a9f8fda1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"df32e8054f7652547ddddd0ea75c6db79d1a857c4311860e2217999cd1d5a2e7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"729edc911bec74ecd4c7129d312eb18efe2a0c69b292fe1828c85ebb2f537539","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"5c0d6bbe7e96dc5827bcd9b6dd10d14ac257d41eb6a82b8b44220d7b4ed6af63","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"01e44f5405bacda4afa3c04deb96bb5fe393e86ba067053ae9f435d3dac58189","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"82401d51c28a8e7bbcd5fbeb1f287c7bac541be0a0a80af0eef8e5218d0b2819","bytes":5258}},"requestSha256":"a81d7d6986a0a77baaaa090a3c381cc00e777b714a4d243c5ef65bf1d95ef2f0","metricsSha256":"c40be7ae2ea285c570c90422e836f8769fc625b1fc282c60bd234ba91c471e9b","continuationSha256":"4902b5729528d16d32f2a4368bffae579bbaa35ed38ff9cc8b3a127a3bb3f7e9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"3da15c04f0c4e8ee6457edd54d27d778be0ff11cd13ee60abbb301c16d957679","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"7ab01b699e7912f3b886c22f20e9de01f8fc9e2ed2f72aab998f65d926b57056","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"5c9912b05a38b4629e9e8dc8128d60c59fbab85b4f582e5aec8d95f169f44e65","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"a6d1719ae88b6d1abe9dadf362cadc3695b2351c26ea0056aa9b4654d04728b3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"f00930f538d1fd449b85c3c1bc20af2d0787e37c4749fa175dd00866a9f8fda1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"df32e8054f7652547ddddd0ea75c6db79d1a857c4311860e2217999cd1d5a2e7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"729edc911bec74ecd4c7129d312eb18efe2a0c69b292fe1828c85ebb2f537539","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"5c0d6bbe7e96dc5827bcd9b6dd10d14ac257d41eb6a82b8b44220d7b4ed6af63","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"01e44f5405bacda4afa3c04deb96bb5fe393e86ba067053ae9f435d3dac58189","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"f9a59ff97a03ff5372f03ac89a5ff481b0e4385e737512a389f5319206a5714e","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"738ac68007a8ee6c02ad29b366f5d25fe1096a4f56d7c54cd57737ef6a39b401","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"726114c13eb4db67a0b450a5336e5e1eb52e3ce013b73a721033f9921c2c9a19","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2481783","recipe_identity":"d42f2c93ff946079f0b6aaec500543cccf45b203a75ab839f259150cde1f39d7","loss":3.5877496646091758,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"bd10e337b3a17fe433ba14617e1bb647a319410ca8fa7b13c171cd3dffe1c5b3","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"bf84fa538548a0e0cdadc754a56ce83a1cb6ccddcf0a98f706c4cb14c9934328","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"4d04d974529f45e2fc15f96b5c10d529435ce34334e68b5f2b2ffc908c569006","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"f11c8d675625776596bc369e98aacb59fe1318857e072c8ee8ec5dc430f82b70","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"89748e040a21e945c1c281dd93847591a2b7b98c47ef756f8013e8ba302a411b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"7002758a36ee386082eded27656555aa93de77650e7451a20b8401fa8de47a48","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"89a6cebd2a19cc2c91a1de59194f498350f63fc32f9cc4b1c29cc6480973cff8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"f11c8d675625776596bc369e98aacb59fe1318857e072c8ee8ec5dc430f82b70","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"89748e040a21e945c1c281dd93847591a2b7b98c47ef756f8013e8ba302a411b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"7002758a36ee386082eded27656555aa93de77650e7451a20b8401fa8de47a48","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"89a6cebd2a19cc2c91a1de59194f498350f63fc32f9cc4b1c29cc6480973cff8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c03974001655591a2894bcaeedae3fc452d4e0bf8c27e0dcf5911d99a04709ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"7415222b54b0b887667d4791284cdaffc6d2b424154812dd7e19df6b85f6dcff","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"59e1714401ebe35737ba0cac3cf3d22ac949fe68d5e2923228d930062949856d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"42fd891d883dc0a08634d9744dd600b1084902804e534f52f1a28ece05b61cee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"c2f5921bdc0bb1090fe101aaa4cd9982184929d82f153f5a3ccebb30b065dc10","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"a80c8eb2b0c8a1130959f036b890f6fc7222eb8efbfbbd878871272ba48155cc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"571c66926ec321f475c9160a74e170ebefb44aafad6b45a407d554359b9cb10f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"3e56a30e786063ae2a7e71bbe2d455ad02aade739b13e110cf639f3f3776875b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"8df1578215e0fc5eb20c78c5b5972762fcb23221edc70de9d043fe03233f9781","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"1b51abc0a24476db23815aa60ff7aa14bb34969b25600c0c054a6c9765470f42","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"1bd23da92329ffc83ab5fff3426bd73318e6e3b70ee393c25bcd336d9786ef88","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"4c36c0b51abeda39bd859e8f2aeb55f7b37cd90381790398f8c9828772f834fa","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"5002ec7996efd313f648ca482d5720970a366154b282a8fb5e668caf335e4bfe","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"a6f794d2f2318240804e4a760ad3a96ee70b95ae7e67c7236580436a1e260613","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"d4caaf2d79d912d3b44ca6e85f6349ce54f2bbcb44b628a7ced0e92f24d1169f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"4d6154acb65784a1fc6ce6aef4b28455018d02f06332cd78843f3a590514d0c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"53f0b6dd4be63646f97c38b8d8555bbd346c673b787c9b5c83c37ca4092c2d65","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"5352160926d0b49ffdb2092f933398acfde7e9cd0e1f0fb156350039bd877a54","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"c1a5fe4659ac0b97a4dc222d889b4b11abad0e9f7f5591e95542bc9df18a63f6","bytes":5315}},"requestSha256":"e0e76decd92d2bfdd912b5419e6354f1687e612552627642d93b575abc7761ef","metricsSha256":"94c1f4e9d5069d4ea2a7f2e5c389ccdc55e0031ee7e3550f8f7493fc1e82b24b","continuationSha256":"13028ecb5fe105bce53f0e0bdaadcd839d3074b7464d6d13f2d0f59b5097abe7","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"1b51abc0a24476db23815aa60ff7aa14bb34969b25600c0c054a6c9765470f42","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"1bd23da92329ffc83ab5fff3426bd73318e6e3b70ee393c25bcd336d9786ef88","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"4c36c0b51abeda39bd859e8f2aeb55f7b37cd90381790398f8c9828772f834fa","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"5002ec7996efd313f648ca482d5720970a366154b282a8fb5e668caf335e4bfe","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"a6f794d2f2318240804e4a760ad3a96ee70b95ae7e67c7236580436a1e260613","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"d4caaf2d79d912d3b44ca6e85f6349ce54f2bbcb44b628a7ced0e92f24d1169f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"4d6154acb65784a1fc6ce6aef4b28455018d02f06332cd78843f3a590514d0c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"53f0b6dd4be63646f97c38b8d8555bbd346c673b787c9b5c83c37ca4092c2d65","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"5352160926d0b49ffdb2092f933398acfde7e9cd0e1f0fb156350039bd877a54","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"e8844566ba9ede64647babde74e10b581b0dbffbfd2d4611c8aa6975a97b7438","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"7a37cc718c440a30a395a6916856fcb90b91372c0b77c5e62151bbb41405be7e","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d42f2c93ff946079f0b6aaec500543cccf45b203a75ab839f259150cde1f39d7","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2481784","recipe_identity":"ffcf85665df13ce512af4d02b8a298aee9a77c498a8d08127e34adfe487f1330","loss":3.5747625272557997,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"6b9330b40e814f9bd3a2be7612c78320609e9d24b115db5ad1ccbbc7ab122bdb","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"34f6b9c5c4e32fa3717213523e17d44a2c8820ee52a0c4e049c19b82df90057b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"c9580f65856a735781b3dda1c3c98ec44ee6ae01af07eb255c4092ae151b50c1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"814bd616c3aa00770b3bbb91e73bdfc3fc173e951cf4c4095c100e7edf73db81","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"fd9d852a24715b589378d3cf1fdb1ec25ae116bfc405d7c3b39507b2377bd093","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"11063d9b8fe281a5824e6125151903ef12150a0e979cb85b3f05ae2bc3e512de","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"c88955fd79bc8683f5b36744f71a34c216f72bcf40679905773bcfa3ae2249f4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"814bd616c3aa00770b3bbb91e73bdfc3fc173e951cf4c4095c100e7edf73db81","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"fd9d852a24715b589378d3cf1fdb1ec25ae116bfc405d7c3b39507b2377bd093","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"11063d9b8fe281a5824e6125151903ef12150a0e979cb85b3f05ae2bc3e512de","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"c88955fd79bc8683f5b36744f71a34c216f72bcf40679905773bcfa3ae2249f4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"642f155005e91243386e8d0d9a9c0090d91ae1412ff76144c1efc8bbf5462879","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"4d167a223af2fc824a6e0aa9ee9504d9f3f4ca397ffb5aeeb2d5d8d0f4ac8812","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"58520601848da0ac41757c22b846e0d0ea961823fa8138e9a729465627e316f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"bb557ebb026292bf287597c2f6ad4f2767796fe4a3a9d775435766b5c8f4635e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"a7d270b19dff0f9ee33ce79733f470dc4f05e3620ba672d8da6a96a74c17585e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"33ef3f6ada49e54853af23bd3cda0803eae6713fd24af1f521134e1af8ad871f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"61647504ae24fe20b733d00e0572acf84163e9391cd23a98277e569e44bebd54","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"ff0db2e39d9575d6b1eaed6936c57a72ef2cc80ec682c04cfaaa9b3750c260be","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"02308f19288ffebb082f43e60d2ee97fe745a0cfbfde2ca7968d13888620af9e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"5328404474c6deff30059f178fabc3f298ae18266e35061106c50c6662cc466b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"fc6a931244119a2fa0df9e156b9c7bb84ea56252670e4d6d07feecffed8c76fa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"e0d946fdc261801b738e80303b8992794a0345bf0c3ad906d36591cafd1a0839","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"530bb16844ad6526ce4497836eaff5c954d3a411059656dfa39ce7f42b80c482","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"0bfd36703c1c16db619156045ae7093967da0b57ca6d78aa340d59976afa617b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"1c70ea6dcbc22a55a7a4a4a544c14069f42f4e840c89a59dd2bd6c19be65dad7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"6a37814531d1d6fdc85146737119eb99a14fcb7a640c613fd88da772d220c981","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"51f91022e0c73f8249323947ba85f1dd3f1f81f6f342bcb402ddeba0dad22443","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"0c61ba1bc00c32c7b125c191b7da7a0bb62a31958a907ced20629085c0995ddc","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"83fb3e1f847e15fe8666e6e70315849282d986592fe0529be12a22c6597d2db3","bytes":5258}},"requestSha256":"e192d021b9bd1e856cff449163572ac53c70ffdca78d73c6cdd2e085de572e44","metricsSha256":"81e6e458e0d41c81250216293c64af94afbf9a7b70e38ad9ac48fb526de746a8","continuationSha256":"264b126bbd90cda8fbc0667ed9151d07d3ccb91833b07a29ba8273d88f1d5c83","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"5328404474c6deff30059f178fabc3f298ae18266e35061106c50c6662cc466b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"fc6a931244119a2fa0df9e156b9c7bb84ea56252670e4d6d07feecffed8c76fa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"e0d946fdc261801b738e80303b8992794a0345bf0c3ad906d36591cafd1a0839","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"530bb16844ad6526ce4497836eaff5c954d3a411059656dfa39ce7f42b80c482","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"0bfd36703c1c16db619156045ae7093967da0b57ca6d78aa340d59976afa617b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"1c70ea6dcbc22a55a7a4a4a544c14069f42f4e840c89a59dd2bd6c19be65dad7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"6a37814531d1d6fdc85146737119eb99a14fcb7a640c613fd88da772d220c981","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"51f91022e0c73f8249323947ba85f1dd3f1f81f6f342bcb402ddeba0dad22443","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"0c61ba1bc00c32c7b125c191b7da7a0bb62a31958a907ced20629085c0995ddc","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"a66425a64f170909b864fabc06db048b3b36c7248970f3c1d092849d640ae4c5","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"634ded3f2f4e15474f6f79f49ceb386ee373806292a187a000a35af3500bc1d2","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ffcf85665df13ce512af4d02b8a298aee9a77c498a8d08127e34adfe487f1330","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2481785","recipe_identity":"bbe93010ec1db7eeb1b69352d308c493b4b2686a6ef81bd08faa0df9e28006af","loss":3.582056787771072,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"2b72f6447143ca19bd28dd318a458f54354ef06b220a76feb748ee358dd2fe4c","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"29919ee8b429627e498550cc44031032e13722f54d2ff3e5e38f0de9ecbaefaa","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"d277158ab02883c59c9edcb40519878e8cb04edbd7a36ed199d5ca4bb8bedb56","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"91b2968d3679194a63b6ef8dcd39a982e9f90093a1a606e01269e0bb65184d00","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"2cd1e13ad7a008ba18377c1191e831e042ac163eb1cc9319359b85bd74ff5cb3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"c15c50755d7fed900b60b571c455edd94feb1691e158434fb108fd5e40206e8a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"b5f32ad1ea6118d12c5217fe836ca25fff7cd0a60bff21d2d1e50b48092ba3d9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"91b2968d3679194a63b6ef8dcd39a982e9f90093a1a606e01269e0bb65184d00","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"2cd1e13ad7a008ba18377c1191e831e042ac163eb1cc9319359b85bd74ff5cb3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"c15c50755d7fed900b60b571c455edd94feb1691e158434fb108fd5e40206e8a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"b5f32ad1ea6118d12c5217fe836ca25fff7cd0a60bff21d2d1e50b48092ba3d9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"305a46ac5bca28636f31c50e20a4ac51c9d240b2e35097c9b83b4011c2859ebd","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"36fb2c069e1d3b22ddf8401098f41da7b026b6547634424ebeca576cd0cd62c2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"ede1b2c2df0f43f3db7e1ce2fb01f204d995b45b74b8b84375a8dec790f63d51","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"116f3b8013651e1b9b2bcb3e292abfd2525bd1313ea36f69a9ed6dc45687cfdc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"1a188ad1a47cb0b381de2ba0f87d5b7f7a1272e260e98e70c175280feb278be8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"54878f7b054db9362f78dc8eae3da9dd1c547f792befc241bb86c227ff2d792b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"0cc132ef9306ccaa591adb92c63d844e33654dda4d9a557360f86fb10df9263c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"27bc78dc71efd8d5b44692728044b943dd84d79de8b05f45797749c1653ddfca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"c49984b878313a2087de127a3738950fb813623ea4c28cd740f1b693c6292d3e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"de0704404627b554246550b4044bf1948e8edb80909aacaa72eeecf8bccb0766","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"2e17d60c69e0c0fa8bc92592c3db9abfaa7f1c811f14520067bd5456c048dded","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"3d7777093c81aed47da2abd565eee9b0a69297e2051d88ef7978c516af1c4732","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"18679e2a5c515ce9aebd05ab0a7b1511b562ebb184bdb5b3d8024bf4f78089c0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"a9ad27de7569faad0160ba9f3fa8c9dcc2de713e5ec7b8564f2f55221fb82a0d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"5ac1d14488ac283e60aa168ed0f81569b7d757fd7c59e8536e783e4622e43078","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"5a86157a2014e0d58502e9d5dd5f7332d9364e189d25e6b7df4bc69664f114ae","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"73581e97bc177ad49b746db875f069bce88959e5f148e21d74227126067c4c51","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"53141b12dca57400562844c7f7584d06e218e67dc6c411b9994bb32e305ce98a","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"550d71a95695b14e319821c4aaa09a9b213a0478dcaa674b681b9bc6233c8228","bytes":5311}},"requestSha256":"def49d3f112734dc25fe9c587dd108c57b596f5d72915d03825e5740f066b85f","metricsSha256":"c11331cf8faf4a14e35d8530a344bc5033c2d0f7ca2cbcfff0d9912fa6320344","continuationSha256":"dfc48c43dce0e0438876bd294e41bab65b6099785fda05e10184baee6a4f7774","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"de0704404627b554246550b4044bf1948e8edb80909aacaa72eeecf8bccb0766","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"2e17d60c69e0c0fa8bc92592c3db9abfaa7f1c811f14520067bd5456c048dded","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"3d7777093c81aed47da2abd565eee9b0a69297e2051d88ef7978c516af1c4732","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"18679e2a5c515ce9aebd05ab0a7b1511b562ebb184bdb5b3d8024bf4f78089c0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"a9ad27de7569faad0160ba9f3fa8c9dcc2de713e5ec7b8564f2f55221fb82a0d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"5ac1d14488ac283e60aa168ed0f81569b7d757fd7c59e8536e783e4622e43078","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"5a86157a2014e0d58502e9d5dd5f7332d9364e189d25e6b7df4bc69664f114ae","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"73581e97bc177ad49b746db875f069bce88959e5f148e21d74227126067c4c51","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"53141b12dca57400562844c7f7584d06e218e67dc6c411b9994bb32e305ce98a","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"c5026b012fbdb5f50ddd699d1fd39a8b0eba857b2d14e6e8d1fc3ec5dda8755a","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e328d55db899d770cc42ef2d7ad0d89d73b6641d9411aaa03965bc32e6db42e7","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bbe93010ec1db7eeb1b69352d308c493b4b2686a6ef81bd08faa0df9e28006af","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s42","horizon":80,"attempt":1,"job_id":"2481786","recipe_identity":"82cb9ebe0a4c528ccca5b3ebd451b922c89d5a2242edfc7dbe62230cfd2aa1c2","loss":3.4527778053217584,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"46d1f8d6b999c1c68477851012b4cf6dd799214269670ae6d1827553f3e7e2ad","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"032a293869909978c1dc17d3243ee1bcd40d6d8be6b8eafe75fb7da35d9d9658","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"3f0c4ec6a9a91113bac026e3a816ae79f2601b53b2e8a96b83364fe1e401adbb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"3e966f257ed9f6888e67bf7d7956fca153d0ca574bf4e3746d9d114c34193db8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"f9e613b484a1e5a50509c25c1d518e1aa6b6ca7be17e94c92becfec385e220f1","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"5db52981e63e73c94be7f1ee641b90317795e2107375cb6976ef3cfec08141c6","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"16ce9ed7a541cb5aebe67b6ae3b7237525a91224963aff7806647f954027f10c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"3e966f257ed9f6888e67bf7d7956fca153d0ca574bf4e3746d9d114c34193db8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"f9e613b484a1e5a50509c25c1d518e1aa6b6ca7be17e94c92becfec385e220f1","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"5db52981e63e73c94be7f1ee641b90317795e2107375cb6976ef3cfec08141c6","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"16ce9ed7a541cb5aebe67b6ae3b7237525a91224963aff7806647f954027f10c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-143.pt","sha256":"f93b6a9c8ad40e70b1be3a1b9153bcb374572ea3aa07e9ef73a5093945465cf4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","sha256":"1f551fe960ea51ed03ec09fab5d4e4e3e6fd13cbc19a04a1f8b89e50515fff75","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","sha256":"1731895882a91be8f9c3f94729a924aef9c4fbf81b31b5a4302acce0600ab7ad","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","sha256":"6c810e395e37866188f60ef515c516dbcf2a7e4de82cd32923cc9c0a7329dfd8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","sha256":"40a4f0f4546a6b34b2ffbcdf8c6e5725f9549b553e58fe5b9d9ef6a54b9df3a1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","sha256":"58e9f317a058f45b7d22ab043f8b9d0c1fc1e21b154a7fc65e8d7640c2ef5993","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","sha256":"5b5a183b69d66cc4ecf87bcae2bb4e98d334ee2ddc798642849d6c00227684a0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","sha256":"0bbb03d010947e2467ea0d3c90ba63c4e71336aa0fa7caf0fc7617f66752feb9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","sha256":"dd39179379739ce3f81fd16eb70f4249835e50354a278d281d1cdcc4b897d8a1","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"ac927c8c55a90e48ad8ba2c8986ce3a37c84b797de6d344f3635ec9ccbf9dd45","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"d70438a82afa959c542412fa66e9fa93d68820c0db6c14d9c18db31cde71b11c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"e30491f62636477e5373ce394c63f08747e9e47203d03189958a80e6f3daa888","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"e71f1092fe9bd0d9a6fd891d6134fd4bca768637691d93431e4c0110e3a664b7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"438e1228856b7f4012dbb1dd32c64a14a7b9b3f84527439c48326c8390c3df90","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"29ca9f773d8c3e7eacf7dc256317b9984823ed1febab2ae6851acaad2819dd00","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"5eb3b0a150549ae2d1bfeb0b79afb2a20420f3ac3ca98fd326ebfe120b43fb6b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"59265fec6e4783fda6577923765b91c4ced9cd0abe3121041eb486b9cad22f3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"f1dee93f24c3f82fac54c093a4c411a70f8a2e1400cbbe4436c55957f0b3afec","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"0d5f0d5f3a0ae7818e5d75529edcf6e8aaa52948d67ee2e846bf68bae80634e5","bytes":5315}},"requestSha256":"482077af01cf11ec4020e5b08e45813e22c4ab3bc671ff90a8c57844713cb3f2","metricsSha256":"f299ad1ddcc227544a7186cf6a233cb7afee5897000bec20663c076dba8c0b66","continuationSha256":"3c82b22f9e05fe1834bf31e2f504c9a940a2f03cd8d8ce1193157169540041f7","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"ac927c8c55a90e48ad8ba2c8986ce3a37c84b797de6d344f3635ec9ccbf9dd45","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"d70438a82afa959c542412fa66e9fa93d68820c0db6c14d9c18db31cde71b11c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"e30491f62636477e5373ce394c63f08747e9e47203d03189958a80e6f3daa888","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"e71f1092fe9bd0d9a6fd891d6134fd4bca768637691d93431e4c0110e3a664b7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"438e1228856b7f4012dbb1dd32c64a14a7b9b3f84527439c48326c8390c3df90","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"29ca9f773d8c3e7eacf7dc256317b9984823ed1febab2ae6851acaad2819dd00","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"5eb3b0a150549ae2d1bfeb0b79afb2a20420f3ac3ca98fd326ebfe120b43fb6b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"59265fec6e4783fda6577923765b91c4ced9cd0abe3121041eb486b9cad22f3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"f1dee93f24c3f82fac54c093a4c411a70f8a2e1400cbbe4436c55957f0b3afec","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","sha256":"910c980bcd29f90be13fc593c46239ce325ffd9349a8220b4c1e7455ce5a4d71","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b5a45cbfc7efed828568898c64d16ca598b6725dcd40838c76102abaa78365f6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"82cb9ebe0a4c528ccca5b3ebd451b922c89d5a2242edfc7dbe62230cfd2aa1c2","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s43","horizon":80,"attempt":1,"job_id":"2481787","recipe_identity":"947e6da995d5cde436e065c4367d604ebf1dfc4d9b1cd3f75b00cd4ac7ca768e","loss":3.4597880102890275,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"1e039def3ba99ed76abfbda2d831f42ffb60fd95f8b4ce553a1386acf5e0f49b","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"ad69f7a1fc64534ca4374bc850fd53ffe75244f7d5747b4447909cc6c4592814","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"7d19a3fdc855c3f41eba2ddd478112e0d1ba873f66a9a2ef08bc8bd09ff140b1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"30ff49ec244de8f7a239a004cf952db5dc2909b131b8a52845e2b14b33749a6a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"0a3604111e15af3f7016878e50b87f30ef7ef31411b6035ef7da7fdf67241474","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"9ef2cde17ef247d76aaf8b9dcb1335205a6430f4594588ea3c165dc187568b98","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"7a6343a37e85a1a9cd558191d079d48e536bae3824dddc828caa4100c28c5f0e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"30ff49ec244de8f7a239a004cf952db5dc2909b131b8a52845e2b14b33749a6a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"0a3604111e15af3f7016878e50b87f30ef7ef31411b6035ef7da7fdf67241474","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"9ef2cde17ef247d76aaf8b9dcb1335205a6430f4594588ea3c165dc187568b98","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"7a6343a37e85a1a9cd558191d079d48e536bae3824dddc828caa4100c28c5f0e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-143.pt","sha256":"d11d0b5993b8e1177bb439859bbe8a8ef30951bf1cc597f81aac3b907f7f60e1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","sha256":"d15b4492d5f32c2a304af29896673bf98f7c635df85b6dd4c2548233194ca30f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","sha256":"c935e8bf5011627002e5e49d7ee4577bc0ab07951b0fd58edaa85d33c1ae2b98","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","sha256":"315599bad0167a608243176ac1be6b083ae89067d783aa1148d15c5484cb47fe","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","sha256":"bd8ddb63e8be73092a069a08f2da45352fea6d57bd13fb576b47a0fbf065ecc9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","sha256":"571e0e2625547263f67b612eb2f1f56f8941c3f02faf0e336485e2a4f6639496","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","sha256":"e43d42208dccc0de9d962f02500358997dd1a35e483e5c0dee3d5f9ac5a01138","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","sha256":"22c3629a2f5e492285fba0ab13d3259a2316118faa8683b18e409073607999cb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","sha256":"c6c144a1081fc069437be8bbe2f7266e301cae7d4e99f5218256678052f7ce23","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"09bb1ff40b0569947279674af56dcc67e7613243de4c62c1bbfd1f774f87d9f5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"1504de2aee8013d9d8de98beeed95c97d0b575e510a7d586762a98f34b2624bf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"e290c7fad8951618168c217bafdada93a2826496b80f14fe46ebaada7ab91c00","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"7d2ca854b37cd638e00ea3154139ad057a5a577a59665b766a474eb8c8419924","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"5f1361c856777c453ee6f49bf88ddf7dceea57d74f1c2af0699f640568eed80f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"2f7d8cdf9f4849c62680b0299c78481f4f53a4a8e2f1364f6a8a06d645c2b232","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"a956deca18493eb6d990267411ee2889e244e2f70e070d214828d4c7fc2c2a45","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"84e349310f4cd21f9105de4de9288af61702756f603faa14edd0a0ad819dc869","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"6fc467e0312680e9000c81bbab8addd34b04be558598c2f7c3d636201a5b7722","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"de2ab292a3c8db6b337f42f9e07d042039db5e9f909f134fabcaec2a911ac5a5","bytes":5258}},"requestSha256":"32e3af9b647970b569d42f1730fc78d01e85f9a1f68734a59fa76fc4e22de601","metricsSha256":"8c32ae75a4f49732c81087a5f40a18eeeb74a17519b548c05c7b90557129e6a2","continuationSha256":"1874143a7ada767c0de6b41a3d6088493fbddc62311eb94ca511e43a1ec777e8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"09bb1ff40b0569947279674af56dcc67e7613243de4c62c1bbfd1f774f87d9f5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"1504de2aee8013d9d8de98beeed95c97d0b575e510a7d586762a98f34b2624bf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"e290c7fad8951618168c217bafdada93a2826496b80f14fe46ebaada7ab91c00","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"7d2ca854b37cd638e00ea3154139ad057a5a577a59665b766a474eb8c8419924","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"5f1361c856777c453ee6f49bf88ddf7dceea57d74f1c2af0699f640568eed80f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"2f7d8cdf9f4849c62680b0299c78481f4f53a4a8e2f1364f6a8a06d645c2b232","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"a956deca18493eb6d990267411ee2889e244e2f70e070d214828d4c7fc2c2a45","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"84e349310f4cd21f9105de4de9288af61702756f603faa14edd0a0ad819dc869","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"6fc467e0312680e9000c81bbab8addd34b04be558598c2f7c3d636201a5b7722","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","sha256":"efad503d7bef052bd222c66fa8a24c9c7450c2bb28772ae070cb29913963fcf3","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a5ac5094a6c0730d756cc681f27e85e69943bb37ad7bec445a8e2cdcd6c60859","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"947e6da995d5cde436e065c4367d604ebf1dfc4d9b1cd3f75b00cd4ac7ca768e","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.006-s44","horizon":80,"attempt":1,"job_id":"2481788","recipe_identity":"19644bae128ed05a53293480ee643f1d5b9982325f699f3cb5f85e6f033c3b87","loss":3.4537285134678437,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"293ed6ae6c5dd9567a926b401ec17f19cc8f926e9a6d43a4bd0f7495b96bad2f","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"09ef96a00bd3162fb731ed0b1f7da9cab102f11de80f68ba1b3d1a1f1a31aeb5","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"e84b8f67e6067449fac24f646f1a996af7623017e0aa2a7ae13a90adddfef1b9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"603843318be822504ba1950f68a7a648a30697c20cbb3c6a2ee6b4561d112dda","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"54f257a9b18efbdfb3c503b54b1b9cf0430ed2b71a937c6619817860197003a2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"e2c93c98551022c014a9153792b0114486a46581bda5037ad32b4536f5b6d990","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"30ddfadfaac78e2c57b97cdeaaaf8a29507186f5c1b0059a57283315b60b333c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"603843318be822504ba1950f68a7a648a30697c20cbb3c6a2ee6b4561d112dda","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"54f257a9b18efbdfb3c503b54b1b9cf0430ed2b71a937c6619817860197003a2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"e2c93c98551022c014a9153792b0114486a46581bda5037ad32b4536f5b6d990","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"30ddfadfaac78e2c57b97cdeaaaf8a29507186f5c1b0059a57283315b60b333c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-143.pt","sha256":"d4aecea7138558aa422c1ce087c946a02fcad05acca079af875260ce71b6ccd8","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","sha256":"232272998afda271a7c51fe2d755df4b56dfffb0fefe04687857884b1525007a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","sha256":"377c81577aeb74d3ac458fa0b220d43d0a856f06b9d9966325192a08a8fafd9d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","sha256":"4486ac6a54433b3cf283e7d48adbcd0d02119bfd477a98acf65ebb0dad879449","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","sha256":"284fdf57ef06eaeae82e40af4cfff03a4307749a713d303079ea7ceca7a662de","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","sha256":"3b80046801092ce0330d17f1601994f1ae31ccca42c4642c6c620975e886ef53","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","sha256":"b2d95531a93e4f66d3c1a95e5edd3eaee12355f3754fa63457cb1f6495c1dd1b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","sha256":"c338b9c104603e5ce27bca424fbfab4e0719c0cd86b4eca6be41c3f4feaf29ac","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","sha256":"2cb5be8652a134fbd46894985335709ab5d1decde7076bd48dcb0f936df92b9e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"39b1c721667169f932ba73630e17f660fa1edf896f5fcfaa80a71c5369c82e02","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"546bc77a7fefe226b815b76bd70cda86ae70089ae7e5d58dc251ffdff1cf35b7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"34c1012be3f1813ea020716b231c02c270f56f8c1774bf86430403b3ca4e54e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"0e96f3c56a8486adeededdc2a6e9dd19ea2061fec03f7744401e9db0fc73f6d9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"138fcd080196e839ba2c335e0600f0d072281e76b4265a6ec2508b5fdd7a64cf","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"f35eb95f2005913f69856b2e3a3c89553295884f53da8633756836aee240f5fb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"74332d4588d0e8ca928152bb633b1fa5b18ee223bf7c581a0ab6e7a4e452c52f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"48b62dae8acc162e29552b3d8ad67675a01f372dfda548c78d29b981524e0725","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"276e29160b3c6fd87617396d409925880b948aebd316da08e9970e191b3dc342","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"7404cdcafeaddad56a1328b6989de08de13e6629e77f179d2a65e2eb596684eb","bytes":5315}},"requestSha256":"9a31438286ad0ebc131b199f98669971ee843e45865799ee35e3c66e459cab3b","metricsSha256":"2f846567763297439ce3bdbaddaefded8194d0e60c30740cb6f48cf8376dcf19","continuationSha256":"46c932728ac8615abfcc64d167383e5cf997ceac77e8bcc948b6819f1c3c27c5","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"39b1c721667169f932ba73630e17f660fa1edf896f5fcfaa80a71c5369c82e02","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"546bc77a7fefe226b815b76bd70cda86ae70089ae7e5d58dc251ffdff1cf35b7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"34c1012be3f1813ea020716b231c02c270f56f8c1774bf86430403b3ca4e54e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"0e96f3c56a8486adeededdc2a6e9dd19ea2061fec03f7744401e9db0fc73f6d9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"138fcd080196e839ba2c335e0600f0d072281e76b4265a6ec2508b5fdd7a64cf","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"f35eb95f2005913f69856b2e3a3c89553295884f53da8633756836aee240f5fb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"74332d4588d0e8ca928152bb633b1fa5b18ee223bf7c581a0ab6e7a4e452c52f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"48b62dae8acc162e29552b3d8ad67675a01f372dfda548c78d29b981524e0725","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"276e29160b3c6fd87617396d409925880b948aebd316da08e9970e191b3dc342","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","sha256":"3f617b26fd303f5065c914bbbe932b59a653f09d0e1e584413777b060dc495cd","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"06cab3017878f770defb07f5660f005af2725ecaac12d93f5119e139f2de5fe2","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"19644bae128ed05a53293480ee643f1d5b9982325f699f3cb5f85e6f033c3b87","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2481789","recipe_identity":"eeda0fcd35d48383ab961e582cefa068c2438442f04840aaaf0c4d3f1f794515","loss":3.4553320907057965,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"5bda878c02690ada039b7204741c94a71b89946db2010a44cfec4ce5de312d81","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"8e44c707320cd11c8adc22552a6eb5535067f1fa289fb4ae95b09d68ca412e99","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"e640062867e7a34d516f4a65be9977b262fccf9448feb7c4c4e40d9463019e3f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"afb9e84e7d288d4322678c8857e58da4040727a016a7ae91404ae07e2565812e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"11a5ef2b48ad8ff33d71ea71826ea0fbd5dbedf726f4fc11e34cb110229dad53","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"d5cafc9a6db7d9f600aea517cdb07bccd05bc9c2a16e00a324fcd786b27c473c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"d967943ebcbc5bf57e5196c18956d6e8a3f36d9065cf294345d0b61a1822c8b4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"afb9e84e7d288d4322678c8857e58da4040727a016a7ae91404ae07e2565812e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"11a5ef2b48ad8ff33d71ea71826ea0fbd5dbedf726f4fc11e34cb110229dad53","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"d5cafc9a6db7d9f600aea517cdb07bccd05bc9c2a16e00a324fcd786b27c473c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"d967943ebcbc5bf57e5196c18956d6e8a3f36d9065cf294345d0b61a1822c8b4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"ae68cac4ee8ef2a42f39bc148001c6a41e2c1bd3ea547348e731963d29e282c1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","sha256":"649377db307794e5985722e26491133b42999e9a8709da5005a7507af298fa91","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","sha256":"511362b8fd3c2198007c3e09829b5dc8c2b30434de1f7c0f940f2c5a3764cae8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","sha256":"0fd308548a0d5f28314e734a636b6a84e9a8899d08ad5ca471a6b28a8cfe4184","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","sha256":"8474fd02688cef8c5184a195cfcf9429227e683082525c18310f86128f3b4424","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","sha256":"c9eb24e03551a2b98a0f1e6dfb7035df7f429bdddd102fccfcfa73eb14ec0bb6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","sha256":"38f65d0f9c15c8ca0d48e5ad31ad906a9cb8c50405dc7a9be1b413ec4bd2003e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","sha256":"2e69390f9d3b1e9cb5fe7144463612bc7ea8e952fd30cb77a65aceace7eebe59","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","sha256":"a4ae16ae1c8a516cc9a4436ef898c670c5ce9ca8adaf0d5adb8927b29d9d438d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"7a3f326294b3e205eeabf5a74a543793a797490719845a61795ac9c08820be94","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"2b4b6bc0aafb88c5ea034b39940ff33c545e03387276d64ae130e45b116e82c9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"f5dd7d3d7127ffdb8b46d7154053a0cbcdb26c9f19111f3a42318eb7f3bbdf9f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"360e807af40e7aca9e7172a313d3702e7e6ce1e17aae82fc7a2914500d387152","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"dad53b8dbfed92ef06dee7cb87c237533d255a68d798640691c172be50785803","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"b6c9efd88e074bc15a9e9b4bedc874a4708a6581f8eba88786fc869813f95f5a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"fdbf1b78fc439a5c0c89b87a3523e32a5812c9ba0731acbede971155af29bc0b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"13cc657dae9bc3c6a513984e382f3fcfed76890f74f57a5b01c08f551a464add","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"961229ad1587dff0c5c2d242e516950d3bec216ed43f2c2cafa59044556f7172","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"1d783cf3bd5da0269e4ebce9f19c30246ca49d591de3ad6b200f93fda02e16a8","bytes":5257}},"requestSha256":"de1cdb50fffbb5f480ca84583e419d087573c66f4bfdc3347ddb3cfff0cc202f","metricsSha256":"770b58e5f8f153f057aeafe3e2608e9c24d774331df7c88a16b5f18bc0ebabfc","continuationSha256":"9dce43d10791c5d25ba0f795eae4bcd874ab80c1d1524e8012a5ac105fa4c65c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"7a3f326294b3e205eeabf5a74a543793a797490719845a61795ac9c08820be94","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"2b4b6bc0aafb88c5ea034b39940ff33c545e03387276d64ae130e45b116e82c9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"f5dd7d3d7127ffdb8b46d7154053a0cbcdb26c9f19111f3a42318eb7f3bbdf9f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"360e807af40e7aca9e7172a313d3702e7e6ce1e17aae82fc7a2914500d387152","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"dad53b8dbfed92ef06dee7cb87c237533d255a68d798640691c172be50785803","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"b6c9efd88e074bc15a9e9b4bedc874a4708a6581f8eba88786fc869813f95f5a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"fdbf1b78fc439a5c0c89b87a3523e32a5812c9ba0731acbede971155af29bc0b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"13cc657dae9bc3c6a513984e382f3fcfed76890f74f57a5b01c08f551a464add","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"961229ad1587dff0c5c2d242e516950d3bec216ed43f2c2cafa59044556f7172","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"1ef09b90d0c3b5e730a61b856a4f1a9a77f7a8e30485ae889de4c8d0c36a1c55","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"40cd3959b6054bc530c7b0da8f9bf2c50c3944d4667e0bcd1f78d65fce15dde2","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"eeda0fcd35d48383ab961e582cefa068c2438442f04840aaaf0c4d3f1f794515","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2481790","recipe_identity":"47541316a0cab77f8ba8b633115d432e8b5ca52a7ebe0ac92c0054ff16a28eae","loss":3.4620790641990338,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"2f47839fe63d50953e76fb316368a2b36d221a431367f38bc941d99857e544b7","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"e1fe41e8dd75c6b8c95e50d32e2d2164cb57291d12caef25e8e37ee825cb398b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"8478daf124509d6ed96f24a0a0cea5733955ea96acaf06032963bb8b45aebb18","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"526599d99e52a783ef6fcb36f187e8d003111266b2302e0cf98336df38c816b5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"15dc076dc9099e681eb950ab08b6080ca10fb60573519759532255b35f708504","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"52fcf79053781f8c8bdb32afa5a31825c1057c29c63c1585bf78876b699a5b64","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"b721cb38e64067fc4580f00c7f874386b6e1fa229fbf6c399addcde14097f9f6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"526599d99e52a783ef6fcb36f187e8d003111266b2302e0cf98336df38c816b5","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"15dc076dc9099e681eb950ab08b6080ca10fb60573519759532255b35f708504","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"52fcf79053781f8c8bdb32afa5a31825c1057c29c63c1585bf78876b699a5b64","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"b721cb38e64067fc4580f00c7f874386b6e1fa229fbf6c399addcde14097f9f6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"e9110f281b81d03440100398b23287a10eae5a88502d18e2615046cbe9f98dc9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","sha256":"574979a4cda4f8d794a8f6f25914d21fa3c4fe77775f9e4fce06e85cfa8419a8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","sha256":"d5ce00f381378c227886703222c0ad54a272c9c6a0288f0199ea8eac57c55d6a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","sha256":"be5d1159a799c7c13eaeb43ec8643d902d465b7af34bb91f8c57eb2d2b60cc92","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","sha256":"7f3a7b60ebeb7adc99c6629583665a745009b564c5ed9240a8935429f58e0b85","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","sha256":"c64c837ecc70d0a142e008423a2220e3936f89ce57f36f31762cd120b99e6b21","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","sha256":"ff087e65014ea0d699f7ede472a7a087658ea31163cf0f58eb43ce672e562b44","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","sha256":"4df6922ec6adbb6b5db6961e69b64c402b1fc2705acb0aa7b91e360fd660c6d0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","sha256":"91ff332840b3b3af88f9992cfd41ea0278df84728a0eb09d2d21acb87bfd9431","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"43a645ca3a9d1cbf379fe302eb0752c12c7eb105527b30b58ab83e2386dfa647","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"58b07a82aef9b3133a892325cc6082763a99fc21982cb61f36c3483c86a67792","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"0ab182eb6c1d1d4f903de8febba184de574a70f62318db9939023a438a4c146f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"3a2173b1d4f6809e8fe6407cf6336c2c9b93fda0de1d071c8af97270570ba4e4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"6322343dc02eb45a4910f155a6077d38f3112fbb4891659b77abd2f411820392","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"8d4c66d58c8b7877fce6eabd9a657e6e7b7ab82402159c2b652e06ef1c92d828","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"aa7459e2b5934c09ef59eeb535cbf6b4e69419e5cb09559320bf4edc6bfa5ede","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"ec4aca1d6d9e5eb66863c0f12a0f740f4eb25e4270e5be98cb96055786279f16","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"4eef365842d8941d25085b6637d80fa9bdabcf016a95bb657c384dfb49bf8c04","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"bd94c8d84e456fb22e1d4ee002f7e49a240548d620d0e669b3cf365a01a8ecd7","bytes":5311}},"requestSha256":"c6d736198fefc8befad56e6b7765b87909af52ce64daf6db748d3cdf2036667e","metricsSha256":"4c48437d76f92a1056d84d0e0c82cee37e8f0fb2c33eade590e3054d3acb4310","continuationSha256":"08150d426b84378aeecf24e0c9dce07b49c522e77045cfb6655bf30615b42ec1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"43a645ca3a9d1cbf379fe302eb0752c12c7eb105527b30b58ab83e2386dfa647","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"58b07a82aef9b3133a892325cc6082763a99fc21982cb61f36c3483c86a67792","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"0ab182eb6c1d1d4f903de8febba184de574a70f62318db9939023a438a4c146f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"3a2173b1d4f6809e8fe6407cf6336c2c9b93fda0de1d071c8af97270570ba4e4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"6322343dc02eb45a4910f155a6077d38f3112fbb4891659b77abd2f411820392","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"8d4c66d58c8b7877fce6eabd9a657e6e7b7ab82402159c2b652e06ef1c92d828","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"aa7459e2b5934c09ef59eeb535cbf6b4e69419e5cb09559320bf4edc6bfa5ede","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"ec4aca1d6d9e5eb66863c0f12a0f740f4eb25e4270e5be98cb96055786279f16","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"4eef365842d8941d25085b6637d80fa9bdabcf016a95bb657c384dfb49bf8c04","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"b9a67f757cfa0c58263bdc4126612404c326f4f60cb4c59904d44b0ca49e38b9","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b7d05813332bd3605d4660ea6c2061638407af8f52f2a9a69a56eeedf1480bc6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"47541316a0cab77f8ba8b633115d432e8b5ca52a7ebe0ac92c0054ff16a28eae","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2481791","recipe_identity":"4aff13aa2e8f6fbd2fc1fb7772a27d262d29a8c333daf8fe43723de5ca5710b5","loss":3.4536922939559127,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"71a4276e5f794cc0f842d27f5898eaf1d3523f059952f13d8677343b0b126bc2","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"49b02adc82d6b60f21629de7c8ac987c43d0046e1b8b8271be19e703986f3a3a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"c6af400eca50dd4033c11ce764f52232141df48b89918b0ed7d190ecba740a21","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"e97e6a0667ff44fe962c635c0c012a0625ed2e2cb8c2ae8896ae4cb0f3706700","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"466a1b271da1f1275e3f5b9970c4a3dbe6223f2fa708f0e1bc3b98391399261d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"29c944583e432894080a3d7d0993761f8f2a7ee2614ef5301b695a0d97e8880b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"00fc0e6788cd5edee4ba2bc7aa870d6d5400aa422248cdcfdbc5d0b8a966d422","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"e97e6a0667ff44fe962c635c0c012a0625ed2e2cb8c2ae8896ae4cb0f3706700","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"466a1b271da1f1275e3f5b9970c4a3dbe6223f2fa708f0e1bc3b98391399261d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"29c944583e432894080a3d7d0993761f8f2a7ee2614ef5301b695a0d97e8880b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"00fc0e6788cd5edee4ba2bc7aa870d6d5400aa422248cdcfdbc5d0b8a966d422","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"4d366f555875ebc3523fa307ab4449cb3fa55e72d1ddf2a5ccda3d86a8ec357a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","sha256":"34a5df6c48fd05229d955f495dd7ad14fba4f9e49ccb9806250570cc7ac915fc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","sha256":"c7ae7f1be50d88f0cc40b3b5a030733b5940bd193fd27078746253ca69f5a2e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","sha256":"2b360a114c959b8b714ad682ae68dc5ddf7dc5d8989806253c1a9e23215851f4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","sha256":"fe9b6c5c5a08380fdac7d80b30d4f6b8e23f0ac2ed40ae78e33cddbac2d0af30","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","sha256":"5c7f18578bb3023ea0ba56ea1577fa1b0de8646e1b198685d29ac6a62015b5e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","sha256":"7e479bbb5a6f0f2e4dfd3091429647c016e1554aa5193243942663f6d7ee6831","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","sha256":"40c4922c2eb97f2b3c13ba7cfd781d77070a87fd0532d517b4c72208bf94cd31","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","sha256":"9503ec06fd504a388d0a4b471f2cb83be1494f9871ec326b97c6b368e1c6ccbc","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"a1417ffb64025a6cec270931d257e33f122caa34e82cff4477ca98a090444a8e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"b8e29cb308cf42b84f169b4b45ee0a0e5b49b4e7fb989b19f2060e9d59f00346","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"403c71f6d85b1392c8f7dc4a180eb3579bfb22c94b1ba3d4f491b0edf7c49eda","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"e9ec8cec0196d00c4664d93f12aba19411b86a6dfd403c2146525915e5c4ce3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"7ef696ec148497c4eace7f900fec06fb6a1569cb8ce207fc9c5a0246007f9724","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"3f42f9772cb11b1ab60956c524f36b8b9e274f70825581972eda9854a1fbc84a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"379ad78a07d0aec248bb63b1129c1538cf62db935d0c1922d1310950b6ac0acd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"35c96897e6778111cbb37bd3174b39951e301d2be22079341254520dab73691d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"a6af5de1b31efef82c684fb95c441badcfc8ca5fbb4b4bfa83aefaef31ff8d43","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"d18d0fd34f395b619971d6b3abf57354d947fd3b0e81718364ac2730f1d8fde1","bytes":5340}},"requestSha256":"f61dadd1abc426f1955881b2192ef60fc42aac65eabf93dbd391a8306fb0e165","metricsSha256":"1b7181bf0e8563e07aa6031d3efa8ec8a35f8a3aa390d6dce7be2d8805ab9490","continuationSha256":"0a939c583766e0e8be79d8eb9d82b35578a1726d647beb15abc9c59efc2b0b9d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"a1417ffb64025a6cec270931d257e33f122caa34e82cff4477ca98a090444a8e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"b8e29cb308cf42b84f169b4b45ee0a0e5b49b4e7fb989b19f2060e9d59f00346","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"403c71f6d85b1392c8f7dc4a180eb3579bfb22c94b1ba3d4f491b0edf7c49eda","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"e9ec8cec0196d00c4664d93f12aba19411b86a6dfd403c2146525915e5c4ce3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"7ef696ec148497c4eace7f900fec06fb6a1569cb8ce207fc9c5a0246007f9724","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"3f42f9772cb11b1ab60956c524f36b8b9e274f70825581972eda9854a1fbc84a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"379ad78a07d0aec248bb63b1129c1538cf62db935d0c1922d1310950b6ac0acd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"35c96897e6778111cbb37bd3174b39951e301d2be22079341254520dab73691d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"a6af5de1b31efef82c684fb95c441badcfc8ca5fbb4b4bfa83aefaef31ff8d43","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"b044aa160649b9fa8d086d879c234a5e4c5737d4898c3e6d8ae682dae46494bb","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b9364d9d515758793624c2b37e80ac05d11c84d377a74d28d30e976dff013e0c","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4aff13aa2e8f6fbd2fc1fb7772a27d262d29a8c333daf8fe43723de5ca5710b5","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2481792","recipe_identity":"5a456ccebca83a3f9e12281ebd63334c1406b131d943e2a788ea5de9c8fa0466","loss":3.463526002061837,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"e06a0e00186c19cb1f26ec398264bacc98817d6dbf74fbc73276b8f210ea6320","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"a6d8663692b051298217e2600cc150dbea5fcdf4e6103e88bdad1cb8019cdac4","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"375779bda477ff7b6459edb7b947d1be803cd40add073de6d19eb762049ab009","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"2cfdb498c0962e10119fdbd4a81a6a9ec02901c56850ede0caa45826a40eb5d8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"8079d1deb7e5860321daeabc37d1f879b5a4cba7cbdceb3735f88fb0953dfd25","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"87851221ea5d7091a91932ce70257ec743a9b657de8ca269d8ee8419e8640740","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"642374f232a0a3ba8a735e047d7d231b905500232a86389bafa9c24ff1aa559d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"2cfdb498c0962e10119fdbd4a81a6a9ec02901c56850ede0caa45826a40eb5d8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"8079d1deb7e5860321daeabc37d1f879b5a4cba7cbdceb3735f88fb0953dfd25","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"87851221ea5d7091a91932ce70257ec743a9b657de8ca269d8ee8419e8640740","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"642374f232a0a3ba8a735e047d7d231b905500232a86389bafa9c24ff1aa559d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"621118079a58550a301710de97806f40ea129291b942e606b9fdd69b1bde2390","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"37d485fcee63e3f3994365ba716c3c7a0ece9799bc2915dd43ba7d5107479d17","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"d7f76a287c46cca73813fa01401c5c82bda9de1801df6d8a8c4deff08a9a4506","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"76958fd0b9aa24f64d9c900a47439bfc2f24ed22914175b99d85cfadf4e754c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"f4c502eeb9d3088cc36a7918faff6f27552c38f9807ee9872aa332290ed0f331","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"1895bc355e9e0e53ba5f22c929c51a9da933204ac89d080f278e5a9fe5eff0e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"2f163794082e7baa3f3658e3b635812c4ab7e27c80b9b93237faec86f574650a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"c03fbc2e9eac4f9ddbb8ace553efad744e0782238138fc440866a5f11d03eb2b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"41ef33ca5a8789632decce90d4f7ab900212137969e0a3a288543850dbce6e98","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"020ddc111194468526875c14a198815803b8c8bd82d2924fa6238658988c0e90","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"228b1bd552a32d9fdc132496d95ab6287148f1a483487f64b13f8b472d0ed5b5","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"c760f067ad61abe20b2260f4ac3b8a7526652611f1a39cea25a4419d63112d77","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"16d313e273bd790138851204cf97fbe39258d303d8fd32996544f879e65d7511","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"6fdc953cc2c55e2cf30f67daeff1f89176d1196a20af08bc24de33249279073f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"bf2180b1ef970e2506a53b1b77220a6dc5b43106f3b48dae18aece583f14ab16","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"86954835ad428fc5ecb500f105a6592bb773b68e65d239f0bebceac53c843519","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"46e5c0ee3b60f39bce2dde7892a391d30fac807fc264d4e46724bd73b218884c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"ac1ccd18e3d5fed1cfba75dd1b0c30358cceb9e3f6a164df6ffa3dcdc44454c4","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"859f1e5559978c063c8563a80ac0cf2c7127155a132fe1b5b37cb1f5329a478f","bytes":5258}},"requestSha256":"2696a2dde314b9ce0c803fc46abb70715212c201eb1fa7472e8fe215a38440f0","metricsSha256":"2af27b87a61d04e8f9379e92a1d91bfd35c1db9bb4bbf351343394af042e3867","continuationSha256":"fec5e59a800e0de264c9c9cf155cfddb145c105caf9972e82153c214f1ff6afc","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"020ddc111194468526875c14a198815803b8c8bd82d2924fa6238658988c0e90","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"228b1bd552a32d9fdc132496d95ab6287148f1a483487f64b13f8b472d0ed5b5","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"c760f067ad61abe20b2260f4ac3b8a7526652611f1a39cea25a4419d63112d77","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"16d313e273bd790138851204cf97fbe39258d303d8fd32996544f879e65d7511","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"6fdc953cc2c55e2cf30f67daeff1f89176d1196a20af08bc24de33249279073f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"bf2180b1ef970e2506a53b1b77220a6dc5b43106f3b48dae18aece583f14ab16","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"86954835ad428fc5ecb500f105a6592bb773b68e65d239f0bebceac53c843519","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"46e5c0ee3b60f39bce2dde7892a391d30fac807fc264d4e46724bd73b218884c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"ac1ccd18e3d5fed1cfba75dd1b0c30358cceb9e3f6a164df6ffa3dcdc44454c4","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"01ce42f2ed6ed80440ff95f341f74bb47d0e101b30db634e405274876de0c4c3","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fc05ae5daacc786e30849162ed387493f61093acacf9a2d241e0813f1d1711d0","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5a456ccebca83a3f9e12281ebd63334c1406b131d943e2a788ea5de9c8fa0466","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2481793","recipe_identity":"d48039310e2d4e202c0eb5b27c4b65aac3f316fb26be1cdbe4439bb97a88de28","loss":3.4653852986364355,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"3fd61eee6b52ca4addb5b5eb1fa11e22f8a6a6e8b294fc5c3ed6d2e901f6660a","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"8b566db7c266f44611a4ca7ab7cc4d161375e9d2243bb7216089dda56ef4ba67","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"ef9c5b7a81e1ed273d7b8f26c395623dafd7b14f48eca37c1ab411a8bea009fe","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"a916ce9a8d330474ef900e3797b657d9281955b1ae64cbf7cd511ce5dbbcb33c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"53c2c0610c8a955261466a5ab8c245760d34c1cbb2e47284e9c2a1dccafdd2e5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"b25449b228db33dcfa32140255aac6ddcb7edf4bac8b6a731bc6bffda116c429","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"c4e608d929210d6e1d6b1dd88a2853387cb3d51faf4d4fd297a5b8b515afb275","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"a916ce9a8d330474ef900e3797b657d9281955b1ae64cbf7cd511ce5dbbcb33c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"53c2c0610c8a955261466a5ab8c245760d34c1cbb2e47284e9c2a1dccafdd2e5","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"b25449b228db33dcfa32140255aac6ddcb7edf4bac8b6a731bc6bffda116c429","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"c4e608d929210d6e1d6b1dd88a2853387cb3d51faf4d4fd297a5b8b515afb275","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"b039cce8d74aff65d1bba0052b640625f4652dd6c1c8d7b157a4fc16261e48af","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"595920d0e91bc7dc32a8e7cf48cd0e721073810de777160e592dd82bc3fff735","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"fb164aaf440b9a136b03dbe629bc1b774d2d307c3e735b6867daf5528e637c30","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"8fe060b7c18d0e2ac902516a62b427ea0cb4feebb73d09d37e80ad5a4f8ca7b6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"a7d52a125263e58838cb8c8934e3e6d9a8747772047670b04214f0ac2bd5e81e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"f9e40adedd8d43d553e4ee07a6f4c674a9052879701d50debcc3c37a58831b74","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"06a821b508465095fc966ca3ab96cc1ed410729154dfa52486aa7a716c8b98c0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"e8a6c589911a301d6291890f33264b9f07c1401bec8d5f55c8793b92503e3946","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"880b54cf3b3764d940bfcf7bcfdf0ca14d76ae740f2a9a48fe856205bc12b34b","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"144dd5dbfcfc29acb45672f8c2906f2c4ad2552c31d76f394943a5ec34122ee5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"f90c194ccc3c345ce203c7ca111114f9b8d8649941d15fd11469156babfe6963","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"0985fcd9e0068bf27bc20194f465ebe1e6afefa471047d9e8af9116241d4ed52","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"188eeeed97b77d111019ee5ce3b87782156a9dfd205e0cc4fa77d8f998df7e2c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"343cdc92aa2e439033512e3d8957fbe4d6a821b302d0ad5432c5167041c80f7b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"2c95cadddfa54b2538cab6f3518df768108707513380e323e26fa673b663b747","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"b868414fd39e7d067204dbe953f6dd1d7883ef7fa5fed7c929e090f221224aae","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"85c5f015cfba08bd6cded1a0cbdd81f071438538e12087cf129b02f4836449b3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"bd4abe78e7439d6e1d64a9c602f89234a07049cf1b178a65f6237bfd1fdda5d5","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"58c3919390db4996ebbee910d7125f409aeb4a94a5e4f7b8058512b3d4fca905","bytes":5344}},"requestSha256":"9ee647e5d99352845fb1233ae06aa9fba5148b5d001e3bbec3510052ada9c252","metricsSha256":"4eb3b86cffeaad730a9cd0f05f9b4cd12716d87d361a9a4ed2f9dc12eb5720fc","continuationSha256":"acd58ed5a18b02e47b6799891d6ffc10156401801b9dbe913cc703d267aeed3a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"144dd5dbfcfc29acb45672f8c2906f2c4ad2552c31d76f394943a5ec34122ee5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"f90c194ccc3c345ce203c7ca111114f9b8d8649941d15fd11469156babfe6963","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"0985fcd9e0068bf27bc20194f465ebe1e6afefa471047d9e8af9116241d4ed52","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"188eeeed97b77d111019ee5ce3b87782156a9dfd205e0cc4fa77d8f998df7e2c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"343cdc92aa2e439033512e3d8957fbe4d6a821b302d0ad5432c5167041c80f7b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"2c95cadddfa54b2538cab6f3518df768108707513380e323e26fa673b663b747","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"b868414fd39e7d067204dbe953f6dd1d7883ef7fa5fed7c929e090f221224aae","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"85c5f015cfba08bd6cded1a0cbdd81f071438538e12087cf129b02f4836449b3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"bd4abe78e7439d6e1d64a9c602f89234a07049cf1b178a65f6237bfd1fdda5d5","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"575226af297fd05b90564dc654a833575eefbb48e7830e08d80b9cdcc03d9ca1","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4e5e636520f55be752973fcad4fc6a8567a9f2b5993a6c550323f9a5d6cb506f","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d48039310e2d4e202c0eb5b27c4b65aac3f316fb26be1cdbe4439bb97a88de28","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2481794","recipe_identity":"58c2b208deb6e68c14ae88380404eb05da069cb948be94254ece10179f323772","loss":3.4570630963915265,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"b3a4a1eb92dac059efa99effa445bb28f83c8f5fafc302ac7cf62048c5d5fad3","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"f5a3a8ad26386c8044e28e03363f2c0ee4c845996b321c8850320985874e4b73","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"b7d9f234dcc82e446a84caa47b8b4414fc1366339b167c6304526da0f7deb2d3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"a08382d7533727ad56821f0186fb14c43a98559ab9fa78aa1141bd0250232325","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"345b9eef85236f1f919ef51243c59a685468e3bbd3482dc5ffa2b692574f1aa3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"c32688015816b88f439a8890d189b8ede81f9691caf1025128e5e47cd52a398d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"f9f47756f82c2568d52b8d2b84472bfec4f84989c4852119ce3d36c68bfade4c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"a08382d7533727ad56821f0186fb14c43a98559ab9fa78aa1141bd0250232325","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"345b9eef85236f1f919ef51243c59a685468e3bbd3482dc5ffa2b692574f1aa3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"c32688015816b88f439a8890d189b8ede81f9691caf1025128e5e47cd52a398d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"f9f47756f82c2568d52b8d2b84472bfec4f84989c4852119ce3d36c68bfade4c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"5ce2c3c00466941c36b424ca9e8113ed04bb017e8eca62ca4054e06d07fb4e70","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"7f5f5b5ae2b65b06a87edb93222ac8c35a65bbd69810fc3d704191e857d8dcbd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"dabac010ad0a9c3ee0ded0d52228549683f2667fbced97328819d5c41fc95e5c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"dd2f1f31622a3fd7d7730be327cf474c573a475cf3eaff99f46c15e3c48d67a9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"8bbb0b714c75bc92a3ededa89c89b8dac99f0dd57ea66f1fd8d3a985de80345d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"7348912896e7e5d1905c11d5cf16abd60926b88b3d1defed4306d54c06fdfa99","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"9100bfb7c4ebce9b6f51bbfc9751f6273cb0f5968c1cb33c5a1b2550be690657","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"f0035f45f3fdc3434b323180ecc277ca5e1a0d3daa598a4177e7e6796057e1b9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"8195f534abb815b7fe97ac80bfb73daa5de5c20c3e937eab5b8fc59a28c17fb4","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e1c4ae39e576aa3746be804403ca02aa5cb8252055ef489fc4d97c52c22bbfa2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"b2fd00f41c28103e718df0551baf7c4ef1e5713b70ec2addfd7a40f8b9c41db0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"b09a478e8d88c8a070d509d7312024d8619c1d067f2c6cd0fe2cef6ff1fb2733","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"fadef1a94be62002bc04e4cfe77cbe87aeb4b6403bd78338bf01604c1eea51a5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"52a4fb5fddfd0b85de684d26218c68218dfa92c544a2ee6bf0988a02747ec490","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"8d8267e87d0b9fcfe8e24be7a9141cc618af60c8bfd8cbd61d73500bb66e337a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"4e9d2999627a2c54705cd3fc4c415861de26a4dc547540a7a59f5377eb735be2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"f44ee2b19a9b37b5aa9becd8712b8597c6328091a76f14a6e4747697324f54c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"3b4efe24b1cc28b9e58f252c52641ce21938ec25231a2d5feaafb6c98f49489d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"de111932712dd8ea4cd6ffaceaadcac1f9cf14b33707c1825b9befaa62a9a5be","bytes":5343}},"requestSha256":"6994bf511d90dbc46fd19d0d016818e52bf5e50eaa02fffeb66c172d770f32c8","metricsSha256":"b0cdd20ccaa771f41e9daeeb3b381618c0e2202b3ab8d9daf6cad52fbe5d4d87","continuationSha256":"acc0a7462ece7f62330f0b0e881e7d9eeffdca828791fa8438aed3bc2906f009","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e1c4ae39e576aa3746be804403ca02aa5cb8252055ef489fc4d97c52c22bbfa2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"b2fd00f41c28103e718df0551baf7c4ef1e5713b70ec2addfd7a40f8b9c41db0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"b09a478e8d88c8a070d509d7312024d8619c1d067f2c6cd0fe2cef6ff1fb2733","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"fadef1a94be62002bc04e4cfe77cbe87aeb4b6403bd78338bf01604c1eea51a5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"52a4fb5fddfd0b85de684d26218c68218dfa92c544a2ee6bf0988a02747ec490","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"8d8267e87d0b9fcfe8e24be7a9141cc618af60c8bfd8cbd61d73500bb66e337a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"4e9d2999627a2c54705cd3fc4c415861de26a4dc547540a7a59f5377eb735be2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"f44ee2b19a9b37b5aa9becd8712b8597c6328091a76f14a6e4747697324f54c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"3b4efe24b1cc28b9e58f252c52641ce21938ec25231a2d5feaafb6c98f49489d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"a08186455ffac2f4edae827e73537a31783f15c71b6b1340e3baf9ed7d5e9bb0","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e534528ed3fc9cf6a874bcff9ee18f430ad364aebf21828163f5e5085f8fb619","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"58c2b208deb6e68c14ae88380404eb05da069cb948be94254ece10179f323772","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2481795","recipe_identity":"03caa69fc2a2a3ad674bd7f4d5f32a722d08e5d5ae9a9be7e5f6ecedfd46db96","loss":3.4822807341736066,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"cd2e4ed5c9d72d1c0c96f60d42dc6638b5020adb702a26228b3b2d06e1f0724e","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"c1974ef4a2f474eb141c0f2cf31eb10b1d7d4df373af8d6788801a885c82d6fc","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"ae0002202972f6a692ecc4316714843dde2c75891a2b7003cc536422de628628","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"84fc2b3bff196c2c510169026cee03fe1b45b3e2728b9a1dc28d2445e330f972","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"f9a2d40da15bb732eec87fb1fe26d761bb34eabf4259690b1f796c76c497075b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"d44c06f23f89ffd546ef0670e5717aa01ef64261676be9419af77db4457f6a0d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"cf9b9fe01ea9b27f48e7c71458328e0178ba1a1c6f5b9b72f92ea9d5135e3a85","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"84fc2b3bff196c2c510169026cee03fe1b45b3e2728b9a1dc28d2445e330f972","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"f9a2d40da15bb732eec87fb1fe26d761bb34eabf4259690b1f796c76c497075b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"d44c06f23f89ffd546ef0670e5717aa01ef64261676be9419af77db4457f6a0d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"cf9b9fe01ea9b27f48e7c71458328e0178ba1a1c6f5b9b72f92ea9d5135e3a85","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"3499cba5c2f9330d9fb4fa31df05ef692e2fe2bc5fe30411af8606f7730d6a26","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"b8c2f65a4effc4e48d4c3299f6fb54dc0d750dfddab4f8badf5e25bd432cf721","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"9e8cd05f0eb65d7e0968ea431247cc4542956f4dd21835724861213dcce033a8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"90d9f01cda12261fd69ac45fea2131c3459c10a8ffc5a665c1222daf9eac03e4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"6050e478705ce11540629697de21ae2a91023f1724a42aec032f5ef16681851e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"b3b62e97f591d74f61fdd90734586cc9fcd489bd284dc8c5f0466ea20c851d1f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"30044516a5d45e27a95e3939762cf58d0e96c6e770df89c858374bbe59455b97","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"d7ba5ae6c994c4780d1972f6c1818edc815dddc31081b0c1b69a41c4a0c6c3a0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"ce27c2dc9a88c701e2715e2e658a4cc795696078fd61d78b741bc51025610894","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"17df17c6a50e1e50844a62a45a009cc70cfaba31fc92a8b5881ddbeccb9b8102","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"4971b6e0fa34ad3cc35bd26fddf3f8043e93020f4ddc51602795a5ce3340f961","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"5e6ce7cd03936ad8aec7ce33a5777d8fe19e3eb63ea132d2bdda52e5cf4dc8a6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"00941a36623cdc2a4a0fc2989d8433002839774e08a5011eb3f4cfe798efbe04","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"9d28175b2178ff31a81a33547c600069eed29e992613dc690f992dfa65eca131","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"fbd27b1a22c871aadeff75aed5342a455cffedfc8894fc43a476db0afa90a993","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"172abd25c46ad700c9feb4906e28b4ae4574dd19ef8840d8c6b7f3857d126132","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"7881b0e86951497e5e3325456a4e3d701c278cb1052a12ffdd949c951c418711","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"a46a8dc5522ef8e5d389a9fccd407e34a34214ff0c4606d8fd3487f9185153bf","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"29605b688ee5692b1a15f71770e6b7e81d2372c38b8cb01a2c904a25262b2155","bytes":5258}},"requestSha256":"c34a82204a9b53a6216f23909feb33e97c22a20ab890f609b4bbf0f64957faff","metricsSha256":"95b8876c2e7ed2a886040c1efd35f3d6cb39f4cdf86088a8580a0e3ead06bc55","continuationSha256":"49a877db7b7e5cbfaa2bcd1db8f05abf25f6a1b69cde9487e3c33367954bcd52","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"17df17c6a50e1e50844a62a45a009cc70cfaba31fc92a8b5881ddbeccb9b8102","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"4971b6e0fa34ad3cc35bd26fddf3f8043e93020f4ddc51602795a5ce3340f961","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"5e6ce7cd03936ad8aec7ce33a5777d8fe19e3eb63ea132d2bdda52e5cf4dc8a6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"00941a36623cdc2a4a0fc2989d8433002839774e08a5011eb3f4cfe798efbe04","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"9d28175b2178ff31a81a33547c600069eed29e992613dc690f992dfa65eca131","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"fbd27b1a22c871aadeff75aed5342a455cffedfc8894fc43a476db0afa90a993","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"172abd25c46ad700c9feb4906e28b4ae4574dd19ef8840d8c6b7f3857d126132","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"7881b0e86951497e5e3325456a4e3d701c278cb1052a12ffdd949c951c418711","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"a46a8dc5522ef8e5d389a9fccd407e34a34214ff0c4606d8fd3487f9185153bf","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"72ac01c1123fe72b41bfbfedc9ca1035d869efc99ac619110cfdf94ae7204fac","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"54ce58b2f5b617c15cc60e48a79fa4097dfbcc3f0580e39822985c0dacd72a0a","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"03caa69fc2a2a3ad674bd7f4d5f32a722d08e5d5ae9a9be7e5f6ecedfd46db96","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2481796","recipe_identity":"944a04c9a3f189941e4794e5483996f108182266d8dbabde0450f8c758bdcf10","loss":3.467179883755508,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"adfda0be8dd4eb4057cc181f00b948bd31d24940428a35008b0b127299ab6848","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"004d11938399965ae63b31719c22a7565dd54daa65e559db5de28e7a8a73c6e1","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"05e286607b092b4b4949a15d38e416fcfd269392ab14ce2fbe4dd1ea6857fc81","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"f1f145264ffcd79057637a50eb96a6a639570c628bfdfee03610dc0eda2e3954","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"15cf59c121f64b119da2c835791ec578d309373768bc0600cb7344fc7effd843","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"9c22475d3eaed464626e1b4e825e2c0cf0ffdfdac88b8a3fd1e29791bf99a585","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"96ae77b89696535c14bc08e650ea5c42152e0146aff059b956142e94d18ad7a8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"f1f145264ffcd79057637a50eb96a6a639570c628bfdfee03610dc0eda2e3954","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"15cf59c121f64b119da2c835791ec578d309373768bc0600cb7344fc7effd843","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"9c22475d3eaed464626e1b4e825e2c0cf0ffdfdac88b8a3fd1e29791bf99a585","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"96ae77b89696535c14bc08e650ea5c42152e0146aff059b956142e94d18ad7a8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"ebcd9fbb7a36576244370e23d8ddd78e68145966768664e7c01efedbf1caa391","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"36797eb8de0f6f3ff78304c9b124cc56b4f67d0ec155af95d6e69d8ddb5b7482","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"049c53631d9531a32c0a36d6160206d829d7af47426f06f9f59ca329444154f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"6be3da0657f93e52db58df4e2d089e3b24ac5c588e602ac4a04cb8fedf97c27f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"8fb6e061bfb2b550a71ba548b7662ff497c4549df86492109dc3042ae9bb5322","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"ea6001d6faaf7d7a831a676bcbbaadcaa9b81025798e1d1e1fc88a638b6bb88c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"5971765f65f4c7e6f52b35818ad0097af56eb21d47f2d4616c5404c3eb48ee3f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"528e1422712accf98808433fdd0166b5a95e564ecc7d6da05379c064199c1d71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"d35a4757b6f4f6abf17b0c94dc5868bfc2e85b41db449d7bb7fa1ae292b4f363","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"bb9841ab941a52345f3a0bad004084d65d59aaec60201f391d34678c2c3c3deb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"531410faf5912ecbcd3b19322c42318a3752caa59da0f5f96ce00e5a27741442","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"0b2af87d0ab434f1943eccb459daaf405ce75d3ad6c94ff90741ff48d82fb859","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"78eaec4906f0dbf1f31c4be488d7748963435388099b8e3cc0883db1958c61df","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"9c5635b5867a5795d5765b4927982652076d0e7eae1d41a2da24d98f0d9ababc","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"6ef7a2ca20c8eda232435101953148d9a4c6075ac86fa380ae76bcbfc5d2a808","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"3cdf4be6e1fc155b76a9e8e35d9cdaa9efd72e0d8a0d0747b43c80c7cb09e06e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"6a7eadc220d6b2b90b222afd73c8123d1e1bb43adf81b3451d8ff9f560f99493","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"8cba7cd4190c46dba94e1f3e95d4bcc31a97d5fc8a2d327e65a81f77f2e0fa47","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"0fa8828879b3b4432ef9d15f68a92d5514e77b35c438859c4915040e64131c43","bytes":5258}},"requestSha256":"b0560d6cbf1f49ab4055b2a39da20496196099c3c7974d54539a0e8b3374554b","metricsSha256":"4329a831f9bd46c0bb77b08345a30e02a9a3ea9b008d11029967fd39d694653f","continuationSha256":"68e888cef56f2db6590ea7451df873514455832bfec14c99aaecf940b2e7245f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"bb9841ab941a52345f3a0bad004084d65d59aaec60201f391d34678c2c3c3deb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"531410faf5912ecbcd3b19322c42318a3752caa59da0f5f96ce00e5a27741442","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"0b2af87d0ab434f1943eccb459daaf405ce75d3ad6c94ff90741ff48d82fb859","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"78eaec4906f0dbf1f31c4be488d7748963435388099b8e3cc0883db1958c61df","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"9c5635b5867a5795d5765b4927982652076d0e7eae1d41a2da24d98f0d9ababc","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"6ef7a2ca20c8eda232435101953148d9a4c6075ac86fa380ae76bcbfc5d2a808","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"3cdf4be6e1fc155b76a9e8e35d9cdaa9efd72e0d8a0d0747b43c80c7cb09e06e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"6a7eadc220d6b2b90b222afd73c8123d1e1bb43adf81b3451d8ff9f560f99493","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"8cba7cd4190c46dba94e1f3e95d4bcc31a97d5fc8a2d327e65a81f77f2e0fa47","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"623d7577ec1fae2d26a7eb282807b8aacec9a4e7633ad5d40aaa11fe6b654fb2","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9713dd1d21764c88c0284f32a64f5d17fe5f01b1ef3e433bada2698ddcf2410d","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"944a04c9a3f189941e4794e5483996f108182266d8dbabde0450f8c758bdcf10","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2481797","recipe_identity":"9530a1585cd21cfb097163755126210cf89e860866d192875e180c1e015b13f7","loss":3.4738058380975665,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"dac7e6338b2b99d45f84a3b22a250124be8610b8fc1382dcdc478595797ec1ef","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"b2b0c29aaf61165e960a9b36ed0aca682de04b16a170719f27fa8676e5dfeb94","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"44db6205e125a97478a2bee29ec808a7fc48c478826d870d923b6268f0aa5e94","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"e00bf28d5ba5244bdd0735e6d547596e74dcfcc5f79b1beb855714c06703e988","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"ad04c347d137d17d3d0e37566e74e00d341cb24ecef384dc29f80d3666af9d42","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"1b25bd1fb2632700ba0dcd56d897a4e03c7ca83928d90c38b4d542a467f43a64","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"5fe6a607df604de5411dc30d3e037d6342e1559fe9ef0e1eedbd7a8f449feefd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"e00bf28d5ba5244bdd0735e6d547596e74dcfcc5f79b1beb855714c06703e988","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"ad04c347d137d17d3d0e37566e74e00d341cb24ecef384dc29f80d3666af9d42","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"1b25bd1fb2632700ba0dcd56d897a4e03c7ca83928d90c38b4d542a467f43a64","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"5fe6a607df604de5411dc30d3e037d6342e1559fe9ef0e1eedbd7a8f449feefd","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"dbdf510e8d1e3c1bdda2d2bee1a1cdf63c589138718ffe75852d4265a21524aa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"cbacee768f95f6672e1ecddbb6eedca14d40b6cdc66e163207ebfcf0126a96da","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"b9c19d348f2982aca0542bf5d317677ae50e0826d121c55e2c60923d539335ad","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"32c44c3bfecd9272912460acbf8f160c46631f6face525f1f709aadcdf540106","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"99dabc2a1e7eb79b9b7ab05bbba379fb08a724fa51a685d227cd168ab59532bd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"a750d0dabf0cb17fe84d2339ebd05ea61cb98dc2cf152654d3802dbff3bd34cd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"432601e8ce9fb9084a4cef33babed2525e947d00d43d44fb10036f6d36732084","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"960c587412df79ea26020efcea6db8cb71a7651c87d3b87afef9b1bc8d2a7a1a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"80969b0147861c3658970d9a5f61ad014001f837ad5a71109d963da164919de2","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"4ab70673699adf4a77c809f8274abecbe32332542098e3fcd78509dc7db9c237","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"3c8ad2d650997db8958e178d49229958e15d8e434735fce5f8c32f83f768acc0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"a293f6c745c89bc00e8585118b3ba7819653d5676d6baf5caf7e2e1be88df8de","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"9f1d2037f6b8f84e1c2b6552dd9ad10f469f8787392dc3363b07e86309d975cd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"69942f7a5a5d58aaf665ea0382223c29f17d421bc4a7f52c8e2568e2ca8742b3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"3f081d3d9f926ca3e554fe88d73919ce571bdd969f359e0667f1e8cc20435bf4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"31df002e2c8d1cdffb3eeed962b386c38909f0cea6201dcbb3646cf314b817bf","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"348dc11fda40cce52f5b7fc8356856e22cfe757aba41ef2546b3ed8c378f1e3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"eb2f224fc64f41941db6e186bef54462b0406e7750a565718cd806bd0052e344","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"b86a1422f223d0685c3f3a712114887501fb9e8403564d3d3ef4c1c83373a0b6","bytes":5344}},"requestSha256":"6e580336842b349535a29d77c5c335bb6f1130fd578fa927d3b53006a8a50d69","metricsSha256":"c0f95dd401525c30e8549c4bf24862801afd5ce360cd96c6e7746b0242d34acd","continuationSha256":"1e2d4e2f1d0c314ada064d47c183f5bcb74e8b837a8b8760d744b5548b873754","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"4ab70673699adf4a77c809f8274abecbe32332542098e3fcd78509dc7db9c237","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"3c8ad2d650997db8958e178d49229958e15d8e434735fce5f8c32f83f768acc0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"a293f6c745c89bc00e8585118b3ba7819653d5676d6baf5caf7e2e1be88df8de","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"9f1d2037f6b8f84e1c2b6552dd9ad10f469f8787392dc3363b07e86309d975cd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"69942f7a5a5d58aaf665ea0382223c29f17d421bc4a7f52c8e2568e2ca8742b3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"3f081d3d9f926ca3e554fe88d73919ce571bdd969f359e0667f1e8cc20435bf4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"31df002e2c8d1cdffb3eeed962b386c38909f0cea6201dcbb3646cf314b817bf","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"348dc11fda40cce52f5b7fc8356856e22cfe757aba41ef2546b3ed8c378f1e3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"eb2f224fc64f41941db6e186bef54462b0406e7750a565718cd806bd0052e344","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"0479e9f79b1b56cb5a4fe01dc4b4e00612ed50d91208fd8cf155415e3600705a","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"cc0b0fb43a4ef294aa2f9a0ffe5cd07bfebccafbba86f0fcbad43f26f35f10f7","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9530a1585cd21cfb097163755126210cf89e860866d192875e180c1e015b13f7","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s42","horizon":80,"attempt":1,"job_id":"2481798","recipe_identity":"14f600e3d2cec74d65fdf5d93a86e0dbaaba3e13a9671b2ab8defc723c11c468","loss":3.415323920062671,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"34f3824a135c2c08c5d050bf8be6d60e47fba3a1bfa3cb60dc583c9b75b9b15c","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"fe3dea8aee6f6a9137755882147962dac52a04daf0d26a93bd66d1a1214af2f8","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"cf78ff2fbe9d9d9733073e2f06bbc3264a565837bccd9dcc5163dc54526d1377","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"330170566e685c20864172c02a3c3a93e5a6d6943a05b840efacbd33cbbbde23","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"b5caf5612fa3d6e9f914057843e5b17597dda288acdb6dc4f0af68a550b153d2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"4e320f4a1082234886dbeeff450dc97a159748687d6ea69671fcd3a8f7b91354","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"272d7f9621c9240005c9742995752c97e2d8859b03d809960591d87f578b57bd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"330170566e685c20864172c02a3c3a93e5a6d6943a05b840efacbd33cbbbde23","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"b5caf5612fa3d6e9f914057843e5b17597dda288acdb6dc4f0af68a550b153d2","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"4e320f4a1082234886dbeeff450dc97a159748687d6ea69671fcd3a8f7b91354","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"272d7f9621c9240005c9742995752c97e2d8859b03d809960591d87f578b57bd","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-143.pt","sha256":"694f75e30b3061a7a6985899a88349ad3ebad99cc419dc2fcadce5ab21594104","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","sha256":"dabd8f429eee93e43c4b242f5794bbd4d0b4c6108ae0dc84741e8fc32ae32d6e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","sha256":"7b1c43ee743a0f817f7862613150317434aeef91eff6814bc831d7b2cf61ce5b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","sha256":"c25168aca39bba7db0bcde5c013efb858b71a6f468c59e2c0e1f05ca366efe0f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","sha256":"a7f77d5ab2177d607aabf5ab7e564e7e0ffe1c44b4fc5eb18acebb0487a48c20","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","sha256":"5e82d9000fd3786e633db8a72cef1638dde7408d74ac75742ccb57f8b29d0e8c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","sha256":"4d69e274ba33d83a54b1ca1a1613b6883111558188f9ba969f2292ef4226c827","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","sha256":"1fbf63d6152594f45fca55ca9c3c88165770de42eea5af6c383b16a38b1d1ff2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","sha256":"fc62fbc5b719bb3335dd80c86eb4de414e400016b03dc9c3577c46d111a06dfc","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"fca67c525afb55e374bef1477eeb45f0b5d6d78f71b17a2736b2f34cf87d7071","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"ae014dc46eec7af34e32b8c891890e1e0682a6edd5303231d1a22d8ac8d8e25b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"a46506a3ee4d916c48a80e8d44d428b194cd41d608ca325826db5320b914452d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"ad21b9feb8d3faf85ddf02bf64046a693c62a3cf02f2d6bf44db0c430417e71c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"48242cb1a01f2b8618443d43980b445acb07460e9e4e09719ccf62222748c916","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"0f86550cf10b9b11b013b866115d22d5f828273ef105905acf2a1f859f6a1982","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"4f9ff87890c62ce241f8e7d4fea6ee8f2903245181f22180ed65cc81385dbb87","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"debdca2e1fd0d2f9cf64afdc17e4d0dd79b07a346bc1654e115afbedf19a7a4a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"b75985113ce9bc45e296f2539fd342baa2d8d38bacfb7f303c8a6b3d8f933d09","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"0818f49948815df4531f3fc0d743fd4af33e0fea23b504cb4f98984b2fdd5e7d","bytes":5344}},"requestSha256":"870be0940cd202049febd0e8ca2d573510a8fefb77683f3e5ffbc296192a9f84","metricsSha256":"3ebb0a53a48173fc505df1d9757ec3e2bb561c7266ebb82e329c45ee438f8aa2","continuationSha256":"4987791e052faa86379f4af800d352ce5763fb56a23373f061ad7f209ab64c58","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"fca67c525afb55e374bef1477eeb45f0b5d6d78f71b17a2736b2f34cf87d7071","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"ae014dc46eec7af34e32b8c891890e1e0682a6edd5303231d1a22d8ac8d8e25b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"a46506a3ee4d916c48a80e8d44d428b194cd41d608ca325826db5320b914452d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"ad21b9feb8d3faf85ddf02bf64046a693c62a3cf02f2d6bf44db0c430417e71c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"48242cb1a01f2b8618443d43980b445acb07460e9e4e09719ccf62222748c916","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"0f86550cf10b9b11b013b866115d22d5f828273ef105905acf2a1f859f6a1982","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"4f9ff87890c62ce241f8e7d4fea6ee8f2903245181f22180ed65cc81385dbb87","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"debdca2e1fd0d2f9cf64afdc17e4d0dd79b07a346bc1654e115afbedf19a7a4a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"b75985113ce9bc45e296f2539fd342baa2d8d38bacfb7f303c8a6b3d8f933d09","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","sha256":"ec38ccccab8809b2dd2d36706608fda344cf75271fd322290e62cbd4b3e08d37","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8c904158e05ead49242cd34183b9d21a57fbde80b95f3ffdf67ebef3de44ca62","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"14f600e3d2cec74d65fdf5d93a86e0dbaaba3e13a9671b2ab8defc723c11c468","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s43","horizon":80,"attempt":1,"job_id":"2481799","recipe_identity":"917572a4d28e11c125d22aba6ab6f5346b195679f6e052afcffde207928f084a","loss":3.4161165450787174,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"f0d52054ed744792d080983254c10f86b0fffbd9ac82aed4244090764d08aa1a","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"414b8c699962142814269888082b9d28d9ca2e15c49c60422f131bd870cf5cdc","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"05d08b21b227174eb00465fa84e1eec8b24e1887af483976c0b783c75620666c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"4d256b04a4e5c480eb78216d04ad8fa822bc906df59dd225ba14ccfcfdc032ae","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"501252397c7e2e5dd4952e253aae5e778f3981cc70bccb02115a74fd976c7b27","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"69518a96778326d0389e9cfc990627f3310c9546a5fa63922fcc2314c7005bad","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"e127dfc7bd7d4b538d4a3561661b9c9ea4fc3ee652c6003b7cf5c989814cdce8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"4d256b04a4e5c480eb78216d04ad8fa822bc906df59dd225ba14ccfcfdc032ae","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"501252397c7e2e5dd4952e253aae5e778f3981cc70bccb02115a74fd976c7b27","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"69518a96778326d0389e9cfc990627f3310c9546a5fa63922fcc2314c7005bad","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"e127dfc7bd7d4b538d4a3561661b9c9ea4fc3ee652c6003b7cf5c989814cdce8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-143.pt","sha256":"e916cfcbca4e32917dfd70714339bb94fb2f790f1054d1b66a14bf346d7b7088","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","sha256":"20536fdb3e7dd004a901bb0495d5f76566a2c203c21dadfd72516d7344ea6d4e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","sha256":"f4952036a111275e560bab669a9bc51968730e31c1b4288d7c5f02ff8dbe8602","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","sha256":"58012161156ea2823b8ad1a8e7e49f1ca32f13f905bb7b9ea21c2c50cccbbea2","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","sha256":"6b7645d6ff2224c21aa646f8fa45ef1d4ec2ceaa8b5ffd59a8ea4e98b0c7225e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","sha256":"06573e5326707bfe2a6b53c2564d3876fc2e59538359de290fb1dd76573c09dd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","sha256":"ae0a908423b116044ece2dfcedace521aa6d60ec0832fc2c5ab23538b82daa69","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","sha256":"a1719e8e6728b7471d4ee5fbba8c95527da45295c91848d1ed86e86d1a426f58","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","sha256":"af42d2f31beb4b09c6a714622c4c7fa7f7f4a15809b678e050a6013b444262d3","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"d59b094d15069e43c76f81fbb3f7799c3d3aed3e0774bc76468c547694fcffe0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"45b10f36c34ba6f42baa221a7c02c93d2d12302a2e874a4efde2b874fa2ba23c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"8df8d9a5ddb292d59977f7ea1e6d02089fbb15e1ae05cfeaa7638a3fe2040737","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"755226eb7ce56f3884bb44d388b357fa6853bb92b4df40a759bc7319ddf14c55","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"e5e233d89c90373db53e4b975f5649c3bd699a236d378e308352d3c39bf32c38","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"e5dae0202d28b2171c5643822686840e520225700173f337a8a1b45ed4d5f328","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"d58e54b8928d4add2b0a207c7e4859095a3a68150730a944d387d54d2eee946c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"bae5330d92ee2d17ca391ffb9a1f33636b47b64bfedc063f46e6ed01c759c520","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"25a3ceea413a0b0254ce45f5df5a77cec1710284aaa3603890824a18b7c483cc","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"4dcd7f1e2c9d5dfba29b24e9e61f6696eb11772f3cef99ed2cc648b5c0951116","bytes":5258}},"requestSha256":"65c1d12339a179d56da7c0460f82e48e15ea82782636325101d181a615f0539c","metricsSha256":"c41e2917ed426f30c78db9d5f4de275c50664780e13d8fb0c3f7e62023a2f48f","continuationSha256":"85952c97cf6752061558ba4c7634299cfa5ee6b0ffda8435e14834ca6731bf88","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"d59b094d15069e43c76f81fbb3f7799c3d3aed3e0774bc76468c547694fcffe0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"45b10f36c34ba6f42baa221a7c02c93d2d12302a2e874a4efde2b874fa2ba23c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"8df8d9a5ddb292d59977f7ea1e6d02089fbb15e1ae05cfeaa7638a3fe2040737","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"755226eb7ce56f3884bb44d388b357fa6853bb92b4df40a759bc7319ddf14c55","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"e5e233d89c90373db53e4b975f5649c3bd699a236d378e308352d3c39bf32c38","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"e5dae0202d28b2171c5643822686840e520225700173f337a8a1b45ed4d5f328","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"d58e54b8928d4add2b0a207c7e4859095a3a68150730a944d387d54d2eee946c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"bae5330d92ee2d17ca391ffb9a1f33636b47b64bfedc063f46e6ed01c759c520","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"25a3ceea413a0b0254ce45f5df5a77cec1710284aaa3603890824a18b7c483cc","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","sha256":"610d137f51da021e45cb7d219204b4377d3fd0145c407fa089b9d0800521eaaf","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b1d844e65cf0c89fb7933824a07ec7becaedc61aa4025ab5949b13be3cc491e9","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"917572a4d28e11c125d22aba6ab6f5346b195679f6e052afcffde207928f084a","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.006-s44","horizon":80,"attempt":1,"job_id":"2481800","recipe_identity":"81cc338e98aa7d5de419f75428f68a57f4fbdb6e7a95acd23d865b20d1a5eeea","loss":3.416442924180625,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"becf9bec97f60662a5e4a2001a0d216f02545a59bf0efff954fa9caec5dd50b3","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"9182680b79a34c39690301f74145a0e7f25fe3fc00e1a05e7b0da2c95985183b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"8830878dd43a36b8079632f98e020600261b8ee3b119a4ee03e68e2877f7a295","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"7c65a845e1d4ba7699b37f06d001c87c24b85eca2f7c70f9a1b8681bcbea3dbf","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"ea7cbcd7c13ca9fca66f24526cfd2b747a94dae8ca1c423bd1873bba4f889b88","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"c0ed107193bec7682700eda14ce8a39022ab5c3b56b1e182778d01928304f691","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"e7a7a7858cd090a9489cc1c1fcc610eddcdf31a5e4526a17e59b4a7f9cae3970","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"7c65a845e1d4ba7699b37f06d001c87c24b85eca2f7c70f9a1b8681bcbea3dbf","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"ea7cbcd7c13ca9fca66f24526cfd2b747a94dae8ca1c423bd1873bba4f889b88","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"c0ed107193bec7682700eda14ce8a39022ab5c3b56b1e182778d01928304f691","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"e7a7a7858cd090a9489cc1c1fcc610eddcdf31a5e4526a17e59b4a7f9cae3970","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-143.pt","sha256":"0d65dd7982e8d640117f3e9b5bce74d257e352968ecf5fbf16ebc2bb7936f95d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","sha256":"4c7f213747084121b2acb27a472c3c9c9fb5004d1c20780cca2eac9e2ddcb4fe","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","sha256":"d584a6fc0ea084ae4ff5a82b557475bb856ff1e29b2a7649dea57669e6835172","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","sha256":"9984cc15ddc6aaee04401691bf4cb964231e2128e37ea870bac38efa08dc7526","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","sha256":"46c0f60fba5e4612559bd2db22f6a845ce4ead4b1b5f47198e96b4f9ee7462c2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","sha256":"799a78041d5393281d67bf9132a946876a68b8d43b67a7b0e6ecef35559e5e30","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","sha256":"400692539fd14af1abfd5c1ae0a7d83fc02d32feb468d2b77bbddceaf0525918","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","sha256":"e6b194016b477759dda8cf10c9a795f885a0e1267af1af4d56539f1f56442928","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","sha256":"ade3307928f92905f16a4f3458539a0cb60143dc0fd930d82fafa91b2168315b","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"15bb3dd088a3e8c2e657b3e6e2243392f3e968940956d20365c54112a5c0bbec","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"7fe92f72591774da2c3d6b18a662766a8a689f101c9317bb2e03259b129f5729","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"e8fc92d491b29208be903fd11c21fb91829ff3767d0c341a4f64a69f4cfced19","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"5db29c39ec667928e5a82746eefff8228040e793b3362036c71e2899623948be","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"48d3f1e886205a9fa5ee6c8862ecd45ec5592ad087d2a61d1267cbbb820a0d01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"ef68c0ecefc817ccd47d2075d2017609e25b4c4834332df5c6691a1f6d628c24","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"939686fcf144175e9c5233026f1fb9b087ff9ebd5cdfa86a9e7f4e5dd63612ba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"5a1e9e03ffdb6efbe51b074542a62b3b743c4161b4c0322307e9de493e217d9b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"10f27352c43260d5450c732a55e9f58ece28dfda81a34e36667a9e8d8931d5a7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"60f14a64880db45d334925347c60f396824b805e19a7c73688d428477448f862","bytes":5315}},"requestSha256":"512a592ae5b1052b4b579d5b8b844fd22942de3879911c30f6d964d2db1c33ae","metricsSha256":"e0278368701042f5aad90a214df5e80412278d8c1639de345d046e0e5500b4dd","continuationSha256":"64295d58b19f52eb14702a822ade814ec68646fda5fbdc385c69582bc6a4f16d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"15bb3dd088a3e8c2e657b3e6e2243392f3e968940956d20365c54112a5c0bbec","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"7fe92f72591774da2c3d6b18a662766a8a689f101c9317bb2e03259b129f5729","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"e8fc92d491b29208be903fd11c21fb91829ff3767d0c341a4f64a69f4cfced19","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"5db29c39ec667928e5a82746eefff8228040e793b3362036c71e2899623948be","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"48d3f1e886205a9fa5ee6c8862ecd45ec5592ad087d2a61d1267cbbb820a0d01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"ef68c0ecefc817ccd47d2075d2017609e25b4c4834332df5c6691a1f6d628c24","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"939686fcf144175e9c5233026f1fb9b087ff9ebd5cdfa86a9e7f4e5dd63612ba","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"5a1e9e03ffdb6efbe51b074542a62b3b743c4161b4c0322307e9de493e217d9b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"10f27352c43260d5450c732a55e9f58ece28dfda81a34e36667a9e8d8931d5a7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","sha256":"6ab936631914c81e484127943af6defdb56368a7ac215ef27d90c122bd342714","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"364db1f3c6359689033ecddc4b42ed28136ca00140d40ae269c0221606520b1f","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"81cc338e98aa7d5de419f75428f68a57f4fbdb6e7a95acd23d865b20d1a5eeea","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2481801","recipe_identity":"9f24f742bc898be04a074dab0265723bde7f5ca97f37c55c3d71d412d454c7db","loss":3.4165495016981873,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"0d14d302c93cb58ef6575d80ebc48c4a93632edd898456785443e24e46210b0c","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"6d34d896a348dc9c0999a005e826530a5c685074aba79c16e8475ec0da9e275c","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"19524f0d446562dc5712cd858458a92a07677da049195148dd7dcef18dde49e8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"25ef8edce60c615ff8ee744c2c1fd907c456f8b43513613417e852c744ff818f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"239890463cecc800fd216d04bc6a0c30cd2d1d4bd3f1b5f910e32753ab84b165","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"b7a8d51ca3d9d13a7d8813860a8909405819b1f4cc3faeaecf08f3636b6b5503","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"6a708c16d73123b5a667420ebb2eefb14ed75fe66e78baf1d3e83bc24697aa03","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"25ef8edce60c615ff8ee744c2c1fd907c456f8b43513613417e852c744ff818f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"239890463cecc800fd216d04bc6a0c30cd2d1d4bd3f1b5f910e32753ab84b165","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"b7a8d51ca3d9d13a7d8813860a8909405819b1f4cc3faeaecf08f3636b6b5503","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"6a708c16d73123b5a667420ebb2eefb14ed75fe66e78baf1d3e83bc24697aa03","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"2c45e066ea34cbf3eb285ef9165290d7d760f6d76ba6f991bbe11d7035d499f5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","sha256":"c8252309f088211ef1c11c75c8b0679214f4a18e5eed852fdc7296ae6d09f220","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","sha256":"f58dcfcf71fe60bddaf5535c100393b3f633185df3231e3b7eb529ce75c27c34","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","sha256":"74bfb61e95179cf6e8eae7e848fe58c0f0389a157bac9cff04037823e4a2b425","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","sha256":"ac63fdb34b8fe67a0b62c607c214fabc776333acb25082313ebaa1a6fb21a3e4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","sha256":"86c6117a0e2f96e6719dfef5d32b81e2c35c8ba9f8ee9d64d6e64e80c3789d78","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","sha256":"d26fd42bc57fc4252f4fee67079221f36e5d0e10efc8b989f09734f5761e4f56","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","sha256":"d54ca981274241d3c4befae3f653d5e06d3b0e6b92f35ed0328f7c8010cadde4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","sha256":"96d1c3a8172b9f63c6435f78a95d78c8b11d37ce76d1ac045c885cb976e27e36","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"862d2ff8434c29109afe76addd28a31b5024e35e5c812f2ba0a67fd76f39e2f8","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"a1af1116c45fa2267bdd0b0d83ea95815a363d5b8b864d47c8040fd51ce7f81a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"daecf67f9e55487aa8a8f41d113d3a9d820d0c6e704eb6fb9f580a5b9658b61b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"e1cfa8e60e799e143091ea0749f1c86b7db40c90be4f22c8c916c6f8a99aa664","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"8ee746a3c92955c432cee7abf2d345851350d59acc91f04088c98775b8b95f56","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"8e1dbb686e516cf24ed1287f412f1576a34093f247fffc92af3bf87f93c67e5d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"b7619f703cd50461d3c7cffba856ec1c79826409d08847421ac73e54c0048569","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"f50bd495a5d6bbed8a345b912d06f7a91be7e13d3e49f00f8e3d6bdb017a37a8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"a4d3439cf2c3c04ac0c8ebaf41b531798472707bd70a39d666385af0f72725c7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"1016f890ec811b50422339a8d0fb85a7d414324c930c6be225558e3b4e6231fe","bytes":5258}},"requestSha256":"a56ecc6f0083b9e55843707d35adba48f26b42fd7f9ee9b638ede43930e938ae","metricsSha256":"985bf05f2ccceaa1a99875711b2014ee7be9fc0c7ebbc9dd511c71b8503ce955","continuationSha256":"970b802986ec663f569a64eeac2174b8635ec66114c2754644a5b066cd70a456","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"862d2ff8434c29109afe76addd28a31b5024e35e5c812f2ba0a67fd76f39e2f8","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"a1af1116c45fa2267bdd0b0d83ea95815a363d5b8b864d47c8040fd51ce7f81a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"daecf67f9e55487aa8a8f41d113d3a9d820d0c6e704eb6fb9f580a5b9658b61b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"e1cfa8e60e799e143091ea0749f1c86b7db40c90be4f22c8c916c6f8a99aa664","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"8ee746a3c92955c432cee7abf2d345851350d59acc91f04088c98775b8b95f56","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"8e1dbb686e516cf24ed1287f412f1576a34093f247fffc92af3bf87f93c67e5d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"b7619f703cd50461d3c7cffba856ec1c79826409d08847421ac73e54c0048569","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"f50bd495a5d6bbed8a345b912d06f7a91be7e13d3e49f00f8e3d6bdb017a37a8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"a4d3439cf2c3c04ac0c8ebaf41b531798472707bd70a39d666385af0f72725c7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"d3d0d0dce8719a7e9ca58ca3228717222f3ac2d83f5bf30c818450e50f0d9ad9","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ad987437e6b33209382bdba23710e07151e0b22463653cafc755d92b2c162b16","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9f24f742bc898be04a074dab0265723bde7f5ca97f37c55c3d71d412d454c7db","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2481802","recipe_identity":"1ccc5059b9c7666d0515f2714395af256149610bbef62f61e6745090033db110","loss":3.414990330896003,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"56564309b493efe01bae202201f41ee87c52b0568b9d1be64afada9502119355","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"5a426933db2cf76ac901aee8ea3be795b410a3fd17225632293b8621e690e354","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"8c1d8a83d0c828334a2557a6cdcc46f601092ca84426d74959c5f94c457b89b9","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"e3fc288aff472bf80d9b412b16aae209f93dd83483702df7b44f8711654f8d96","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"0a52b909900fbc153d3f7782c0724e8599c1b8a53e49ffdae505a9cbc422a126","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"58549293d7f5f896e05aefe21717b24aed70bb1d0416967f02a44186a3f3209c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"e73e6441c3715e42cbeb220ad7021e3d504ca7b902c027e4551fba301607d81b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"e3fc288aff472bf80d9b412b16aae209f93dd83483702df7b44f8711654f8d96","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"0a52b909900fbc153d3f7782c0724e8599c1b8a53e49ffdae505a9cbc422a126","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"58549293d7f5f896e05aefe21717b24aed70bb1d0416967f02a44186a3f3209c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"e73e6441c3715e42cbeb220ad7021e3d504ca7b902c027e4551fba301607d81b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"0939101351974c20a3f173d9e91179a5d7b40866e0596377a927b5e8c464259d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","sha256":"8eadcdd5fa3104fb909b01ddfb846f41067ac69ff93c37a2f334d1ba3fb4088d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","sha256":"9f968f3ec35d7a130489dd9498af49988d9caaf2cd7dc98a84368c4f5b69ea64","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","sha256":"6f37db78eabae68551d177a28c27387441efc63038d57c4987749e2eb85fd4b6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","sha256":"438d3ba7d61f8ad8fddb40b8f8f47b1016874cbafa27927bb30a391967eaa1fe","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","sha256":"9b0e3df439d035d3cb744ad1977f54a56049c611299c9d00fae704fa93774924","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","sha256":"003a9f1441c6cbadf39d9dff00b62bceb74c49d80d2790fd722294dcf3e51843","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","sha256":"8782f7532d2e469e3e9eef754abdf4dc8188eb6bfea5a83026e800ea6460ad3d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","sha256":"e4473cdc57bef497b98c16bff11f03d138629ec76e01517589075093616609e1","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"e2c684069a82f94d86025b1f59bf8a895e094827cdd076e83b63c8c55c62f07b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"d19bedf8016b3d3b94fa92f05e4c1a157ed05ebcdf3e8b6a0861ed941039c849","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"c98d1d6b40f641d6bd285dfeb482ea99f528aafaa8f8c8cf0e85a66a7addb406","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"262976cfa18a29388a8a4f14dc5b705cda6c26524acb32176f1f2dc97b6387cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"79f27011e12286380596c77d8a2339708c887d8b9d2cb43b2d7efc7958cb288a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"a4df207e466ceff6fb97aeb860b74b8c9896eb40989b796b7368763f08ea7c75","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"e4e10271098cc1a8f1419d16b0b7832e6df36ae0875bf99c24421abc70ae5980","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"738b5732bf7cf497c89507d2b8cb0022091d1a629050f1d7eb1704bfdc6356bd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"cee1ef8f020f32db81fba881ab92e836581abf29762206b2aaaaa095f51b55a0","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"445b61abf2a18b57621470fc599c7d6a5cc71b1a340bf771f689d8233026ea0c","bytes":5314}},"requestSha256":"6981e54d209d71c55802018679625d35a86984fdac1874c218d53f1e11ff9220","metricsSha256":"e61071841c138ee8b29589b52d501edaaad66b1fef49946a53e2fd5283b8044e","continuationSha256":"9c1873d1e8fde0f1d43637d9001601fe849805a1ec6e1260e84748373b515452","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"e2c684069a82f94d86025b1f59bf8a895e094827cdd076e83b63c8c55c62f07b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"d19bedf8016b3d3b94fa92f05e4c1a157ed05ebcdf3e8b6a0861ed941039c849","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"c98d1d6b40f641d6bd285dfeb482ea99f528aafaa8f8c8cf0e85a66a7addb406","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"262976cfa18a29388a8a4f14dc5b705cda6c26524acb32176f1f2dc97b6387cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"79f27011e12286380596c77d8a2339708c887d8b9d2cb43b2d7efc7958cb288a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"a4df207e466ceff6fb97aeb860b74b8c9896eb40989b796b7368763f08ea7c75","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"e4e10271098cc1a8f1419d16b0b7832e6df36ae0875bf99c24421abc70ae5980","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"738b5732bf7cf497c89507d2b8cb0022091d1a629050f1d7eb1704bfdc6356bd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"cee1ef8f020f32db81fba881ab92e836581abf29762206b2aaaaa095f51b55a0","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"4fd35c3d8ed20e42707757a5e95027d4544ddbebb1aa62b645428309e283d3f0","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9b653e8a1d33580b8b0293b5e799565b12348e69e1d31ef18ac5f0fd23d2f424","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1ccc5059b9c7666d0515f2714395af256149610bbef62f61e6745090033db110","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2481803","recipe_identity":"68a11f13fbcf0ff1bf047d2ec63d59cac44b09edc8127bf7fed909cfc826dd8b","loss":3.4110390871190117,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"1960fdcf43b8430dc6550ef45237c562f22a499b25f69b47c2efda4e549afe0a","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"1cabd3d0000910b9342452453b51cf1058feebadb973ed0d83feb9d73a0e4b70","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"51f5cd8ae1c4065dcaeafa4ebed21ec958e01dba735247f730bcc7e03bc2c317","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"43ae9d0b30b3572355a3d4b8c93fa8b52133fe09b04ecd787376717ba0f415e0","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"ae0bbc832c1e6717b4cca82f459d020fdb17c237b404dc101a8399fd7008a1bd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"34a066b2228cba9bcbcd25c284027de60c57cab01fccb2c6de0364578e4eefa8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"69d2774e9383d13e85fb194fc589ad6ce0578b132b902f6af10ded9288219240","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"43ae9d0b30b3572355a3d4b8c93fa8b52133fe09b04ecd787376717ba0f415e0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"ae0bbc832c1e6717b4cca82f459d020fdb17c237b404dc101a8399fd7008a1bd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"34a066b2228cba9bcbcd25c284027de60c57cab01fccb2c6de0364578e4eefa8","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"69d2774e9383d13e85fb194fc589ad6ce0578b132b902f6af10ded9288219240","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"c20603b158975d3e18c03916d783b76afe4d6ce2c6180b3f7fd3d8200c086141","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","sha256":"dbe12649c92fed4abf91cb40790725c2e7b36a2bf9e9973d72818d9b6eb7d0d2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","sha256":"ad68d591c5eb72455619d99d11034f27af8e6d6c5af6dc2ce30ea990f7043e85","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","sha256":"a07d425bb1bc88aedfc3495024df9169f69f594946c5bf63e43c4b0e101eba2e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","sha256":"41ee0bc4bde4a68c7503be57df53ca484f729e1e843555a57b389a49507d70a1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","sha256":"65bdcc0c0fabf6329f46f10de368befa6835697fba50021ccc23cc18a68ff4bd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","sha256":"66b860ee4e882a7870a1ebcb8935e09ac93d8b5ea4443002a74cf96db4250197","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","sha256":"9f9d55ed2dd712d91ac0d4a2cc3a09f2c276190491506c2116686e74f7f5bb87","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","sha256":"d5df6d7df2ff9a64b46eb3e0d0eae030ea35d41668b0e371dc6cd2fcd91fcecd","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"168684e24c299867c742b0a70aa06974e0c7df59e8861497c1c4b32acb796a6d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"b3d47151983ccee454e5ef33497d09034b7c99c47f4e86f3770b8611ac1ce5d3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"6cf290c3e1410d1375727635af2847667277cb87f3fc4ecf2b6fb58ee4bd2448","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"1e259f888884bb34f22da95c9fa7d697b015d5aa286de6597e2104ff6fd4147b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"152e83359e9c71b2ad340a759c60907bf3be68f6c4e189ab6d282902aaf6c7bf","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"4e82c8ca2da3b837edfff423b10165002bb38ab4a0aa697ca8e9c04488b1870f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"e7f2dbf169c9bbd6c3dac343cea72b114ad81e4ee312dafcdef79bb790262e41","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"624ff865d00162b4c58f9bad7420dc41346976eee3d78c3d207856aa06d6ad18","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"1009cbc67e0a3fe25c5ccc86dc2fd4b8db6207ebefa23ea278ca2a6f321e0c90","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"d4d0dc5ad03f437bd1c41785ed9737d957ab39556dae476d3367179a0edeb7a7","bytes":5343}},"requestSha256":"2cfef16a39476e89516917c9c364b23c99dfe489348fda2c21965c0c3bed46b2","metricsSha256":"b2a738a51e6ae28a8132576ef33049effaba8601a7fd9482bd45da3ebc106913","continuationSha256":"da99a94f7dd0e073da0d4a1b82b7558bb0af1b1446f5f17b8ebf77b59eeba01c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"168684e24c299867c742b0a70aa06974e0c7df59e8861497c1c4b32acb796a6d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"b3d47151983ccee454e5ef33497d09034b7c99c47f4e86f3770b8611ac1ce5d3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"6cf290c3e1410d1375727635af2847667277cb87f3fc4ecf2b6fb58ee4bd2448","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"1e259f888884bb34f22da95c9fa7d697b015d5aa286de6597e2104ff6fd4147b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"152e83359e9c71b2ad340a759c60907bf3be68f6c4e189ab6d282902aaf6c7bf","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"4e82c8ca2da3b837edfff423b10165002bb38ab4a0aa697ca8e9c04488b1870f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"e7f2dbf169c9bbd6c3dac343cea72b114ad81e4ee312dafcdef79bb790262e41","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"624ff865d00162b4c58f9bad7420dc41346976eee3d78c3d207856aa06d6ad18","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"1009cbc67e0a3fe25c5ccc86dc2fd4b8db6207ebefa23ea278ca2a6f321e0c90","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"0c4457e66138c945caa00a859cdf07e1a8e09a0b8a7c8ed179e72ed81f18495d","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a9945daf6cebba4f76234027768a16c6d468523c119b4a67217f540300a81739","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"68a11f13fbcf0ff1bf047d2ec63d59cac44b09edc8127bf7fed909cfc826dd8b","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2481804","recipe_identity":"e642983b4093fd618803c2e6cc1d3c6de534ae2e0cc2cb09f1c1d30c06aa728f","loss":3.41746784427521,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"1e98032b166c539cdda00dfcdd74eba65ef1e89799f851f55dc2fb75e4587eac","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"20bf6d88c9dfc3d2e1267ca0df1c2f05f547851b331c998403414e33bb51215a","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"9f8a2411901d4780ed2c26b4a38af55780647cf043bf47871f0b7505c0c91d24","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"6a7ffa6e7f68aa772f0ca6f8c744cac2ce32e4bff2a468019c473b2950fa6870","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"53f9be65a51bdac2aa707a4c066a79d7c0fac34b3c0bb18f975ba4d6b278d122","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"3986fe1dd41f1ba5c67f6eea15ee71bd19381e60f7595e18606b9cf692db415a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"6f79efd65959a569601c9a5cb90e9cd012e5169015f12619b5ac5ead5737dcd9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"6a7ffa6e7f68aa772f0ca6f8c744cac2ce32e4bff2a468019c473b2950fa6870","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"53f9be65a51bdac2aa707a4c066a79d7c0fac34b3c0bb18f975ba4d6b278d122","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"3986fe1dd41f1ba5c67f6eea15ee71bd19381e60f7595e18606b9cf692db415a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"6f79efd65959a569601c9a5cb90e9cd012e5169015f12619b5ac5ead5737dcd9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"8529ce3c33b9c148b1a1675c6e357d3d57340659885fdf8d47f4e25a43de370b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"9aa77056e06d0f71f4b164e72400b9220d45232ad23cb763e3e3321924ae83cc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"8532b19baba781fed8926708c096fd57233e1018a847a3120bbf733e4c849400","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"d583d9658ca47df05466eb8a04a4e959bc2b762242caec0de4ab43d762a02aea","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"e266d0bb9b95a130c5139c2a90a04cfc9c34592d9438819d1f0ee5c2790be792","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"9cf119769c4aa79d5d880b99d0612b37bea235dda71bfc18dd1ca45e88794dbb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"ede226e60f857af1bcc21bce704e6eb99fdee0baf565e2ed476babd0927bc2d4","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"53feac1538429baf1cfc0296213859fa7b574aecb59d5da3eb928484dbdab5c0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"854b3f764d9f23ecd2098f8bff58b70a1c21bb2a1ae909494a1c0c61dbd9566c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"98cc427a85c7585b3b896b49edd8bc6d7f19cd88cacc5d5fbe2ff81e86fa8b20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"fb9c48813563e750ed7ee6d9fa76fec927594db318986df8e806b5ae37c798bd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"19519ddd02a1303e99eb7cfb0b7199f3d6f7a5e14c7e07a7023d220c52ebcef8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"b8239339a07f6682f4e7ee7e928f0dc0e4559020ed30011709f2cab70baec3f9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"88d97ca5a99e2fbade5cadd88137830a6c406ff9fef704a499c71d197da01f95","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a7425b46fe17846ab76a9b9aadeb70636fee0a95e94940fefa958010a9ea3291","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"9339a7e91f37f3cb53bb62ec39eb831e6f50e59538efd586c4f5f0faac0a151d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"c7e9967dbc87dea0035d09e0d458c337ddcaed29ad7062d477fc427daf9b37bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"a31f1952a803d1e239fb831d22ac2cff705bef5b1c3b7230befad5e5c4a38f81","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"02972c6300d38289310012dae930190e4266c161ffe0b6f7a7cf4da89de36b2d","bytes":5315}},"requestSha256":"11b375995204aea6b9a894c5280aabaa66ca9e8775977d5b05a8bbc104564241","metricsSha256":"b8cdb92f3fae3537203dae54bdb913129e2b062bfabafc59676eef5d9ed0932c","continuationSha256":"b114c1e4ca2691d38c6750e0b51ce0f95a726808f374462f2692c5d77d513859","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"98cc427a85c7585b3b896b49edd8bc6d7f19cd88cacc5d5fbe2ff81e86fa8b20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"fb9c48813563e750ed7ee6d9fa76fec927594db318986df8e806b5ae37c798bd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"19519ddd02a1303e99eb7cfb0b7199f3d6f7a5e14c7e07a7023d220c52ebcef8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"b8239339a07f6682f4e7ee7e928f0dc0e4559020ed30011709f2cab70baec3f9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"88d97ca5a99e2fbade5cadd88137830a6c406ff9fef704a499c71d197da01f95","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a7425b46fe17846ab76a9b9aadeb70636fee0a95e94940fefa958010a9ea3291","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"9339a7e91f37f3cb53bb62ec39eb831e6f50e59538efd586c4f5f0faac0a151d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"c7e9967dbc87dea0035d09e0d458c337ddcaed29ad7062d477fc427daf9b37bb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"a31f1952a803d1e239fb831d22ac2cff705bef5b1c3b7230befad5e5c4a38f81","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"ce426464d8b022e55e09bb15af4e601f0002aa0d4b1a39cad2978e46c5d41657","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"78117f741c30ec87e30424f4532f1cedffdac6668d753b64e01deb13106cf933","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e642983b4093fd618803c2e6cc1d3c6de534ae2e0cc2cb09f1c1d30c06aa728f","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2481805","recipe_identity":"1eb91a952f4cae71e2d946688da5580e324d33a06aca7b85e6cfcea5d424a357","loss":3.4265044628690835,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"bfad28525d6b87d26a3ec431a96ff0190678248616cbbd14e0da7cee2f691087","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"222a193526ef6035d510f07605d64f52e22890036cf2e7614b5a34ae7b56e360","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"3820f02f00d9d78e54923bdc3e3bb5ce60e81634619ece08e6e414122191e836","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"969badf51be6f1dc4f2812b7d4f88052ae3beb587c1a333d714161c310253c7e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"32117b5acbdc8a0c88ae6e736df818d32471f44ded8a3e31b1688bf9e4448cc3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"9ba02755dcf8aa51557780db14641789c69623c40ff5887ca90046513c8dc3fe","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"2822b7a22ae39c7cffbe686d0e836c693db24ac3ae432eae3223cfdc3a99ae82","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"969badf51be6f1dc4f2812b7d4f88052ae3beb587c1a333d714161c310253c7e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"32117b5acbdc8a0c88ae6e736df818d32471f44ded8a3e31b1688bf9e4448cc3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"9ba02755dcf8aa51557780db14641789c69623c40ff5887ca90046513c8dc3fe","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"2822b7a22ae39c7cffbe686d0e836c693db24ac3ae432eae3223cfdc3a99ae82","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"2ef7234451ad974b27b09ab7288fcde974093f0c1b2df4e8772ad935b671d9bf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e17abf529cc23e2d6f8daf829e5683503d156722fa4aef769a64da286372c183","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"0b6c7ace6148eb8b86a281164c3133a9f94bd9aada2c5d4507208d18d4286cb4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"51cff3f7edf683b6fc5eae1c26b0497b41a572f0e948827bdf47b49aeaa042d7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"43af1a4240dd01040fb9ac2b98f698063f178e617eda2a6e1e004a0e05d7c938","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"b9e059463d2a2657acbf4484d63ec75451343ed579315bb56297f379645d90a6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"de467b96f0786100c2302159719902c00ded203f2ebf94ab6e614dfd020d799c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"7e9be0c0cb8fae030266dcd3cf03d11a7b78b51a347647bc99e9eea1580f3f77","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"cc13c0e6f3e0c3abf838904962bd3790098d2cf9c998e58d5a2f691abf35aa31","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"83a0a311b4ceac8c1d98f98305f6026fe41768351109ca1e0872dfc904e2c120","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"0fc86f701a5613dda0c8d0ba3b1fd227d65593ff9ce7785ed4c58c19c47661e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"3211efde08f882f43455cff389b94b6aaf14f1d91360e3342890a01b22478d5a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"9178b8253b81050c00a7d431427b7b6b541e4582dde5ee4339a68fa2d4b0fb73","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"b60ca255d9ad97b3387da83c23d0ee629762979bb9d8be28f82a5d512d871db0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"59c924883f5321e7392682bb4976bc3660a40054c67317d50bbc7c5edff7942d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"7049eff3a946d71595c71a6df67bb48cff551427b65156e902e43c653699ba60","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"c2910d86accf89eee132261b3c70ba6343587436311b2e4812835e1211b30fc9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"a99288bd7bbcb676046f155e04c21e6569bccf698fe01119763c5946744b83a0","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"98e491354515ec24634fdb86f741041918f549ceaccdfb3afa95aa5b54523352","bytes":5344}},"requestSha256":"2ee504e192c5ccaa0e5f7635d18ebc1c12f8a99c9dd411b2bc9f001cf5e36ae8","metricsSha256":"4d6046765a453b4e95a078132afb98c80b1a7037805fe136ec0c5037166d6b08","continuationSha256":"8c209b98d0454627bcb0b3e9fcbc7ead7ef93ee3afa0d6177c97e2ac2c978268","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"83a0a311b4ceac8c1d98f98305f6026fe41768351109ca1e0872dfc904e2c120","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"0fc86f701a5613dda0c8d0ba3b1fd227d65593ff9ce7785ed4c58c19c47661e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"3211efde08f882f43455cff389b94b6aaf14f1d91360e3342890a01b22478d5a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"9178b8253b81050c00a7d431427b7b6b541e4582dde5ee4339a68fa2d4b0fb73","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"b60ca255d9ad97b3387da83c23d0ee629762979bb9d8be28f82a5d512d871db0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"59c924883f5321e7392682bb4976bc3660a40054c67317d50bbc7c5edff7942d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"7049eff3a946d71595c71a6df67bb48cff551427b65156e902e43c653699ba60","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"c2910d86accf89eee132261b3c70ba6343587436311b2e4812835e1211b30fc9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"a99288bd7bbcb676046f155e04c21e6569bccf698fe01119763c5946744b83a0","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"19e38653ce87d4d33d8db1020c85358e2ab4cd06108d38f4604e75e31aa92c27","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a0d9ef891426920a1f3ed909503ea2591418cb5e784e86f9c3da53cb0b331fe4","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1eb91a952f4cae71e2d946688da5580e324d33a06aca7b85e6cfcea5d424a357","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2481806","recipe_identity":"ea2bab4899467f51e038fbb2f95d9b6be5a71355d311fe0ab7e0ba3a17f1418a","loss":3.4145587232316856,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"fd127c7058c412eb3d8651bbcc540b929213b519aace0f09ba62269aa9bdd914","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"8f9cec472d3731236f1ae025b2ee0ed384b19aa487bd27158c83270ae2ac6345","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"c50aed7602b39ece0a40c3a03e1858ce3e5598774c76110cf398fd478850a4e1","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"ebc8f95cd6c350d2a1b6fcb13648749e6959dccb77bbfa8a60358ec95968aa91","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"9ded3eeb5edac985a132ea6d12b708af44915f418be306f0c5c51326caa9318f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"6d4a093ff81e682cf62ce134aee6502349ab5d82edbe266584c04fde41d59f98","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"5868a799ed1d492298f557f363d6f04f84a6d2a49ded363554036a0205f35557","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"ebc8f95cd6c350d2a1b6fcb13648749e6959dccb77bbfa8a60358ec95968aa91","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"9ded3eeb5edac985a132ea6d12b708af44915f418be306f0c5c51326caa9318f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"6d4a093ff81e682cf62ce134aee6502349ab5d82edbe266584c04fde41d59f98","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"5868a799ed1d492298f557f363d6f04f84a6d2a49ded363554036a0205f35557","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"9330d048862fcb13451a2e9d8810242475a70ba268c3ec78ec1d9d29ff7c6717","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"d0639562967ef175cbd7d7f31494585f6486f3b0f8617702d0e6f6e91c55a2de","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"c96a7418c37dd4799e79ec321e54216d919d14be691876ec0ed1bf85d0d209f2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"87b61c00191af8702c13a8617d457df72089890e0c18e86dd369f0a53b8c1aca","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"9507afe2be1f8f6091c1d5765619b5265874289a960e950056cba3e78218565d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"24dc6e6677931c8673a1fdcc7482a2a79c3a5372894dc92225d43b5f29d46220","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"73989b670feae74c16753378c3b20f822df72369836156fd4f339d1b49946f09","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"f63f8070f87e30d8c316cd4365395a4894790bf77ab964105c3393665a6b17be","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"65a125b3306d9807d6e04c6bec76eb806476c4ef41135c60eceb796a1680786d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"7797009be147d6f28ecfce09c7d64a3e2daf78f9f5815cb22007d89f6367d47d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"d04db7389383a4dd388f8ffd75d1570e3c1136a82d7a73c5421de51370560909","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"23d3f35d07029cd0f94fa85dccf30832e0e8abb95216f4ec7e80895a8ac91972","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"86e600c781b99a0bcb01501df9a595c3a7a5f778aa8a47cc34d2d67febfb0079","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"81c8ca1e0862b62a24e5f04fbe7c91fdc7160fc59bba4a6d471e95f224b81601","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"aec795fa31069d34f1da5b82ead8e34c054804b1cdeb3795891d5d0981aa17fa","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"fa8dda9849919660df26d57ce3d8102182f4cc57b902b7980798442d319596de","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"b32a0bf3ebcd963421abf0d79ade18153e4ca35a0cb69e43c81fc6cc77e1b51b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"9efa24e7889f29f3d3a58da61a651d0fdb573278a1cb4a9e3a5b8c7926a3505d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"1a677d17a1d59aaf2ca2c6255cc4e03f60e3f7594f7813c3de9543c6de8d5e00","bytes":5258}},"requestSha256":"bfd2e3c0b5e5aff3af1a22df6b61b8b1176c36df4f2dbf71595bbc5f21cdbd63","metricsSha256":"97003c3d67f638c8bb745e46986de1ca3673cf9eb02f4a95b8a81ceb71284d5b","continuationSha256":"53dd500ebe0d67ac0430df9c089a81ee9522dceb99dfb811fd52859a99c5b879","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"7797009be147d6f28ecfce09c7d64a3e2daf78f9f5815cb22007d89f6367d47d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"d04db7389383a4dd388f8ffd75d1570e3c1136a82d7a73c5421de51370560909","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"23d3f35d07029cd0f94fa85dccf30832e0e8abb95216f4ec7e80895a8ac91972","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"86e600c781b99a0bcb01501df9a595c3a7a5f778aa8a47cc34d2d67febfb0079","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"81c8ca1e0862b62a24e5f04fbe7c91fdc7160fc59bba4a6d471e95f224b81601","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"aec795fa31069d34f1da5b82ead8e34c054804b1cdeb3795891d5d0981aa17fa","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"fa8dda9849919660df26d57ce3d8102182f4cc57b902b7980798442d319596de","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"b32a0bf3ebcd963421abf0d79ade18153e4ca35a0cb69e43c81fc6cc77e1b51b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"9efa24e7889f29f3d3a58da61a651d0fdb573278a1cb4a9e3a5b8c7926a3505d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"444f4399703db6491b86637cb317097fda2f3ed2ceecb0069ca0d8554db397d7","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f2b83f54f951a4da8b44a9dc4be85d6cda32f886c9038286d8cc2969e6545f83","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ea2bab4899467f51e038fbb2f95d9b6be5a71355d311fe0ab7e0ba3a17f1418a","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2481807","recipe_identity":"6a3e866907f811331edbdcfa10bd78447d6b5ec53457ad94cb50b22945353943","loss":3.433559710226971,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"4f340848c5a2503cf08487773913c36743ef60936686edae9231d657c39602a6","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"0871bd18d52be8c234a740bbef23617bb617506a5f5487b94b7fe37c3805b5fb","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"a26fdaaaa1e5bff4e12f339428ce5ab57e7395c81effef0e9286421ab8d43996","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"065239f029e04e8dfdf61de7d5f5a3f4d98509743bde535091d4ee6b2a26c14c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"393e39862e928d7531815b88e3fadd812a9d6fbd4ee28adbd33e4ba267aa1d43","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"d80d0c75a57584a4bfc42c16629f2fe9451a8f9516429ffe61a1c21e3a4f5980","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"d437f5378ef6ef414eea44e5b0dba1a36a190f18ef2df89635bea9990471fa72","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"065239f029e04e8dfdf61de7d5f5a3f4d98509743bde535091d4ee6b2a26c14c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"393e39862e928d7531815b88e3fadd812a9d6fbd4ee28adbd33e4ba267aa1d43","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"d80d0c75a57584a4bfc42c16629f2fe9451a8f9516429ffe61a1c21e3a4f5980","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"d437f5378ef6ef414eea44e5b0dba1a36a190f18ef2df89635bea9990471fa72","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"cd7d47f25a0ae8fd339961218134f497b69cbfc4e261a4721860157098cc6462","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"adfe8778e18962676edfc433e0ba1f113e18eb8f5a04daae2be094e45a9a522d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"262b06cb67ca70657624937ce31daf1a13e51d251c5d9ecdbc47ceff8de55d61","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"3a2061c2c0721befc197af7a965ba1b99fbf9a8e04c715dc5ff08ae938d18334","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"36e04b1a1f174cac5386289c5577d09e9ee459293d5a2d6c1440051be6aa6760","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"9db492cac366a1bd58740d64f964da9edad447c6217bff32ac09b62c015f7e36","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"d592aca80ad4cd59652276674345bd00b3642d288e298fa132e735d51eda4750","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"c6d41d7b825b705f939c0028cd7e520f007993cc81151c3fad01d59e5b1455c7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"f1ef767f4f15e1e42e2a35c6e0a89fca8031bb8972573ccd300b7da2d831a53a","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"6fce26ca4e5081aebad2233cfe58c6e8217107903aa9a74832cbaef2c8e9a8c7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"c25fdcd7ae3d9ddb30f4db3b313076c899b0dc8a3c419ce8beb13b25000ed85e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"ea5f1b13d5108bad873593dddf65dc8c51d71cb4c99fad399d96afa9d1043213","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"b0d5bd045edd54ef0c0309f2542ee40af992360bbb3cf595f06676c22f864c4f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"62f2dc2f6b271df3a2986f0ec6be68f18d43f82fa1b3e1b275d113805bb7cc9e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"c95eb863abbe8e74796eb53235d97f1decb0dd152d7029a7993df83fd9da3c8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"0265924834470eb810b7d1d48463ab202789561bdcd2af3ad8c594b0c0534e94","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"b56a8b6dec3d7fb02ff322f1336bd73219384a9ab83344995023230ad1e46d40","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"28b67f43919b9eaa85ab0472c48e78f69d75254151eb15548cc8a321ee01c4de","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"4af2e42227c9c02bf5484354eeea96cd563a22cc40d07087bee99a0d69fffc97","bytes":5315}},"requestSha256":"2ec42256647287f82619623d692a7b813552d775108460e25917bf90965e4585","metricsSha256":"d73049c8549261db3952d3b9e09e5aa098a2d2a5e5c6d847efa500c9ff655fca","continuationSha256":"fa83739d8c61dca499c998ca8cc9495e4198308d51a66fc8b8b4e466a4e6d632","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"6fce26ca4e5081aebad2233cfe58c6e8217107903aa9a74832cbaef2c8e9a8c7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"c25fdcd7ae3d9ddb30f4db3b313076c899b0dc8a3c419ce8beb13b25000ed85e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"ea5f1b13d5108bad873593dddf65dc8c51d71cb4c99fad399d96afa9d1043213","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"b0d5bd045edd54ef0c0309f2542ee40af992360bbb3cf595f06676c22f864c4f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"62f2dc2f6b271df3a2986f0ec6be68f18d43f82fa1b3e1b275d113805bb7cc9e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"c95eb863abbe8e74796eb53235d97f1decb0dd152d7029a7993df83fd9da3c8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"0265924834470eb810b7d1d48463ab202789561bdcd2af3ad8c594b0c0534e94","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"b56a8b6dec3d7fb02ff322f1336bd73219384a9ab83344995023230ad1e46d40","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"28b67f43919b9eaa85ab0472c48e78f69d75254151eb15548cc8a321ee01c4de","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"c4f9cd64d97b9a877e7650576e9a8d7c8b9441039b6f3ac3a5982458bae0a124","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"29ce3b688613d52c5a5c02076ab29b18b563223abe3993b727cb048382f37020","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6a3e866907f811331edbdcfa10bd78447d6b5ec53457ad94cb50b22945353943","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2481808","recipe_identity":"658c9cc7a95570e5cf725494839ae854f47c13ee2328532a9f591d2f8cc079d9","loss":3.431684296599056,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"8bbd9a03394d9118651de21b800ae48e3433309f36836cdd27cc63e4f1eaa283","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"38166eaf5d5a9e0fea431f290e429d950dc9d8c0f62bbb39d19dc25440f1c0b2","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"95b778cc30afa63e159378612eb58049bf498a32ea693b20845b2bdecafdac9c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"dbaafa46ca39de6cd87ff71eb45d161a366f4020dc94c526cbcacffa9c785dc5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"cc708272824b67aa1bdcdc924148082906112409a64362f7835947c5e9fcc327","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"88b0562d0347d2a3113e6e9ec37ffafb03b9de6f3550b2e934495cf6fe90af3b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"2daf296322935561f6d6a487b322c51bb12485a08486db29939c3b37151e7e2c","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"dbaafa46ca39de6cd87ff71eb45d161a366f4020dc94c526cbcacffa9c785dc5","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"cc708272824b67aa1bdcdc924148082906112409a64362f7835947c5e9fcc327","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"88b0562d0347d2a3113e6e9ec37ffafb03b9de6f3550b2e934495cf6fe90af3b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"2daf296322935561f6d6a487b322c51bb12485a08486db29939c3b37151e7e2c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"bd185dd7c1680d37c70c26ea0ef8a3e519f6a79be88a7024adff7f14fe0cd214","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"05d82496ca0490d86319930b31146e45d2ac1abf4706cf00435d72d75cc4eaaf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"60f02c95429605848d162efa6cce4a41f9dd83deca21e5a2942dfe80891543fe","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"8326145ae6fbc8b786cfbf63de3ecb007a7c0069293a5287ec0c800566cb16fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"c67a3f3eaa356c11c530d6caf96f67ff7c909da65ab4bc99cfd3f57ccf6ac64f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"cda8ce9e3cb3a5cf8ff95db43f5486d5acce747f00fce2a04cc1a3d571829748","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"41fd124d127c34ffff6cf1232c633369c91cfa1ab2a9ae53049b935cd497dc16","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"05234741fe118d0683a58de45adcb5b585ba5f3c4f6369c138ba743f7fe45bf8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"add487440fa278d967d5ca1b484f2e362c7ddfa49edc31532c6dce163bc421d0","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"e569075f0154ee44d0a8e8093ce74471529046958c7c7d916823dc584048c381","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"7c3664eb26d98d7262bdc753fed055716d215c60f200dada914b1c97504df847","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"45f578ff1886a01356522fbc8e757bdf5ad9248259a6b93813501cebf5b0bb45","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"3b47ce762ee36e4f030d2069ccf4e154a2ce0a11f79c8c6015f027dc29523ca6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"7506113143748b9ddd90019c544ec01ffc573cac329944614c48eb5cc31aa702","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"f05dae5041bb630b68a068c062b7b95def1d10260308511b4d40e1da65ac728b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"56c6573e5c0fc2c97ed3e4fb34ead140926fbb8374c29faf445049f6ea8a7764","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"63df0e1a0aca639c08a9a257b21c5361f4d606f157a8ec52fb4fb6747273647c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"12a41d8487121f520ad5c2fd5aac5efe2b33dc09e059ef3d574a84de34109abf","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"5a9ed624ddd6369c2fd3088b096d2e8c32ee13b50a2880aa66e24720b8e14df0","bytes":5344}},"requestSha256":"30e7c2c362937fbe426debfc5befac89bd1a34f10e879234bc8415d122a42fa0","metricsSha256":"c07079327cbce028a9cfc95e5ee4ad889a01ed2d2e4d218dc199c5950696dcea","continuationSha256":"2fe09cd63315329ed209fa0b0e9a885c6a1e664567824022a024f09ba6984cd8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"e569075f0154ee44d0a8e8093ce74471529046958c7c7d916823dc584048c381","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"7c3664eb26d98d7262bdc753fed055716d215c60f200dada914b1c97504df847","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"45f578ff1886a01356522fbc8e757bdf5ad9248259a6b93813501cebf5b0bb45","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"3b47ce762ee36e4f030d2069ccf4e154a2ce0a11f79c8c6015f027dc29523ca6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"7506113143748b9ddd90019c544ec01ffc573cac329944614c48eb5cc31aa702","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"f05dae5041bb630b68a068c062b7b95def1d10260308511b4d40e1da65ac728b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"56c6573e5c0fc2c97ed3e4fb34ead140926fbb8374c29faf445049f6ea8a7764","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"63df0e1a0aca639c08a9a257b21c5361f4d606f157a8ec52fb4fb6747273647c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"12a41d8487121f520ad5c2fd5aac5efe2b33dc09e059ef3d574a84de34109abf","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"0f02a6d438b6398663e3bb8ab734bb3c0ba35566b190477817ee5e68692728f8","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ca2cb3489a93c46ca47fc74ead8a62e12a62a786492db842351b464a5902e599","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"658c9cc7a95570e5cf725494839ae854f47c13ee2328532a9f591d2f8cc079d9","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2481809","recipe_identity":"3b645335dc09d815f1443d8d170c2a61a5eb999899713f8f2d6bf32aa047a211","loss":3.4305317166636695,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"824d7116e2070799e92655598ef06f06a8fd732cd2408e0e43fc2eba37d1eb0e","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"adee5c15ed95f520f4ef5b8b2eb7dd9d4d5eb4ebb79a22e18f8f717a06427dfe","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"9c37566d5c22e6017988117baa687b59db42a024f041762af173556945b4b674","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"466eceb714a68452f24c6e225e578599710c22487b20220403479de5d143308e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"b74ec46c2113967b2a5fccca2e43d8ef75de7001512c6616c5c14f43c7116806","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"98afe5c2934c2fe5063be6b9f3cf30eea5f589caabca341a187c690705fe404c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"f87bfb4597cb33579caeda76daf5ed6afe04a2c702ee978313a694adf3150e40","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"466eceb714a68452f24c6e225e578599710c22487b20220403479de5d143308e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"b74ec46c2113967b2a5fccca2e43d8ef75de7001512c6616c5c14f43c7116806","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"98afe5c2934c2fe5063be6b9f3cf30eea5f589caabca341a187c690705fe404c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"f87bfb4597cb33579caeda76daf5ed6afe04a2c702ee978313a694adf3150e40","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"4d1f8b71f39c233f48c42083d927e1acb129dcabea5df96b151b2a8c224f8041","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"b2753df54eeb3de9b05c2598f326cd2e0c0e9213322064d137c86aac03d6b52b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"a6d3a8e70cfe70cf67ee64e76a22a6468346d099dc956c17742874655fc77ce0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"bf59102b2467031b9c99f29b89be7dd47eadbb06c5c828bad63a875871a44a26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"0ae00b831aa05057e433b94a162d56824b03974adac24e5bb27c3b0aef8e745e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"6ce200b2cf3fc82711db480f79a441b10d21d98f9c98abcbfea6c957a1d1ec76","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"81a5b4977397fbbc9c8a6915c415fbd3029eb20333e68d1cd2861b6f4605b340","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"7d37e9c5acf69e4ec761ac3e118ca8362d31db589beb8fd7c2b18275ac1dfafd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"af8b88eb46d425bc002e221c5c3d1ac9b5a8d25640d6b682f29cf8bb7f2f9df6","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"687ee7e4988d72a59ca5d78a85fc9433e19c59123eb41b26028268b755a99c52","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"0f929310b513cc9ea4b6e1f1d014b827297176fa75273e311972be1310fec6ba","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"1c4c83a7b95b05925fbb984ae93ceb2140df958165152ccfccafe011339395ae","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"e5ed023bb39bbd1d06f92a6fa899abd171e550aa7c827b1c9dfdf0ebe4a0a199","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"0935742c8ac2ad2184c3be8c9b2481211a5b8ca1f4d03b720b60eef036c05e68","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"caa2c7bdf3b876504f354b338c010cd41caf87f22a57b5e97a057f3cfef4c485","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"1d1701f1ccb8a525d3240c8cca49d387ea51227e16c56399b6d5ce4e54605ca7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"3f83fed38a5cd101e3bdf2b65ff1c5209cc0098940acc94d08debbe1c697cd06","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"8a7be783aa8426a7f1e0a106ea163dbd720f0aca03db558f930ab65d48209c9f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"3792aa1199c59276a04e1ed0a4554c08f45eaa5c1744bc87b8e3ed8113e132cc","bytes":5256}},"requestSha256":"7d9b002d307a3e86cf52338ac8372cf5b1d95d98cd290c0516517cdb06452f2b","metricsSha256":"448961927deeb553bf4748782bd8daf8fd44d01e4d881786c356b783ffcd6a86","continuationSha256":"a9a8688c4615be7d2de7085e4cd5df098a19fbbcb2aaf512d44a007984c04813","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"687ee7e4988d72a59ca5d78a85fc9433e19c59123eb41b26028268b755a99c52","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"0f929310b513cc9ea4b6e1f1d014b827297176fa75273e311972be1310fec6ba","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"1c4c83a7b95b05925fbb984ae93ceb2140df958165152ccfccafe011339395ae","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"e5ed023bb39bbd1d06f92a6fa899abd171e550aa7c827b1c9dfdf0ebe4a0a199","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"0935742c8ac2ad2184c3be8c9b2481211a5b8ca1f4d03b720b60eef036c05e68","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"caa2c7bdf3b876504f354b338c010cd41caf87f22a57b5e97a057f3cfef4c485","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"1d1701f1ccb8a525d3240c8cca49d387ea51227e16c56399b6d5ce4e54605ca7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"3f83fed38a5cd101e3bdf2b65ff1c5209cc0098940acc94d08debbe1c697cd06","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"8a7be783aa8426a7f1e0a106ea163dbd720f0aca03db558f930ab65d48209c9f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"ce5e162b7c5b49adc12c099ecbb9ab3a780ad07c5dc00b05ab543171edcf0edc","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"afb44ca80dc09a20f57270b7ff4e65a83385214f3b90f67c75a098b6c7ceeca5","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3b645335dc09d815f1443d8d170c2a61a5eb999899713f8f2d6bf32aa047a211","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s42","horizon":80,"attempt":1,"job_id":"2481810","recipe_identity":"216d668b7ef33c7acb11f9b716395ef05e1e1bc5b2ba298a447fbfadc77291a4","loss":3.491702525862739,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/result.json","sha256":"04a28051cd2d4bc4d7fe4a309867819b35dbcd6faf1cd947d1c45af48e2ee787","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.pt","sha256":"5b4935977668c73e72a8f072a1bd2e3560ccf445109b0f4c028954aa403d2766","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/final.safetensors","sha256":"2bc85ace1ac7f203e6fe0a5fb210cce8f39ed814d2c2036378219d2f931a7878","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"3023ec7def2c6bcc91977fe3575eb87c5fc741627923543bf0d41f64ac316a5d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"1194403fb8e48454231eb49f5db5f91bbb3fb136b5283e3e1b6262bd4c95a689","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"09ce99e71d32e032191ce1a10aff0e01f8a1cf20522654ee49313a3cb4fccf69","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"7a1d88cd256dd46c5d4e40b0d0810abc318fef86de0dca86d319c7351476ddcd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"3023ec7def2c6bcc91977fe3575eb87c5fc741627923543bf0d41f64ac316a5d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"1194403fb8e48454231eb49f5db5f91bbb3fb136b5283e3e1b6262bd4c95a689","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"09ce99e71d32e032191ce1a10aff0e01f8a1cf20522654ee49313a3cb4fccf69","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"7a1d88cd256dd46c5d4e40b0d0810abc318fef86de0dca86d319c7351476ddcd","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-143.pt","sha256":"17893f0cf8bbbdda29fe32f3768affeb1c2c547d6a2e351210971a6ab26ce021","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","sha256":"6768bef8d387c4bd5035efeef0a93f488747981070791500406f2f2f70b0a9df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","sha256":"fab378c8a725030c700d93c8f9c6a74a5f1732e2dd541edc8d6be423772d0f93","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","sha256":"7b81eafa200205c14c2ed8e3a39624b779cd8fbbd93391c78fc42c8f9607c88c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","sha256":"713e1100ae75ff9f4dc5ae36904dfb4ac8340b2b242b5958358f9edd3df114b2","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","sha256":"8b1dbcfcfc84bec500e6618f869953358c23899f7920ec0d40b605d561135463","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","sha256":"c986c5ceea7ad7c8fbb20e7f046a01b5a954e77ada9c3c3958225e8ce37fa28d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","sha256":"5c96e9466db75a61888587258d6de424221653acad75ebe825a0f9b48d56630a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","sha256":"06460416f80397ef8cb6de0efe26ca35715d9ed7d6296b5bfc900a73874225b7","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"e45d59bf25c9058d701b9de9a19888e2575bf2befc9bf5b9a7d44f75cdd083a3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"b95257ed0af9b7b61217ee44f8ad78730c681ce479557485ae96cccc491b5642","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"82a19d5b96351792bfa919b8677868cc110afa0c4aba70f5448af704d77c892d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"7b0107a1cd5420213f4df7c0de83926eeb801e6732d7d5efa27bcf5a7502292b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"3116023035836e1c349c3bb0f96314c8647191f27c0bf10de8e9134dbc7d2ea3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"af59211c190d0e1951a01cf9492c9eee502aa10ed438d5b0fd46547d17962f43","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"3b1f734640616e701b884c181799fc44cb30b22daa5bada2dbc4c4d9ab9e9875","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"2349ff1484fd492ebf664dbb1ea8b80277ab9fe802f170c0aa5bfb9dc83bc5b3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"f3bd23c3ccdd65fc79d8d98ef1c24708dccb1707f071f76c898ace4d3bbc9bcd","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"9e81fc8d977815e68db88bd8bddd4787d778c627329dbeaf00e04b7090de924a","bytes":5342}},"requestSha256":"3ef407cc0d929f4f39f010d3b21cfc2160b4cda3296d555ee269791e6b2c17d1","metricsSha256":"676caf0a82d18efc22416d48651ec5febab720d33aac430b520818ff04587661","continuationSha256":"b19b402fcb3d397bc5f16dd03451ec7ec33a3388ce733916a7079d3d9c33742e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"e45d59bf25c9058d701b9de9a19888e2575bf2befc9bf5b9a7d44f75cdd083a3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"b95257ed0af9b7b61217ee44f8ad78730c681ce479557485ae96cccc491b5642","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"82a19d5b96351792bfa919b8677868cc110afa0c4aba70f5448af704d77c892d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"7b0107a1cd5420213f4df7c0de83926eeb801e6732d7d5efa27bcf5a7502292b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"3116023035836e1c349c3bb0f96314c8647191f27c0bf10de8e9134dbc7d2ea3","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"af59211c190d0e1951a01cf9492c9eee502aa10ed438d5b0fd46547d17962f43","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"3b1f734640616e701b884c181799fc44cb30b22daa5bada2dbc4c4d9ab9e9875","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"2349ff1484fd492ebf664dbb1ea8b80277ab9fe802f170c0aa5bfb9dc83bc5b3","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"f3bd23c3ccdd65fc79d8d98ef1c24708dccb1707f071f76c898ace4d3bbc9bcd","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001/continuation.pt","sha256":"55725e8c15943cc6b4cb170c0858ddd42e5ef248e7c53cdbeb52939bdba71ecf","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"20c76679f4b2d44ca5bd95fba39d9aee80845441079045c5299f7df4011baa13","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"216d668b7ef33c7acb11f9b716395ef05e1e1bc5b2ba298a447fbfadc77291a4","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s43","horizon":80,"attempt":1,"job_id":"2481811","recipe_identity":"95613d0a72af7a660027f745623185bf8d778fad5d4067f8ca41a9272fb0001d","loss":3.495890585457421,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/result.json","sha256":"aba7b2875601301e050bf4728496f3cc5c0a565a3587be48ca90816a6119208f","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.pt","sha256":"d85413c25b77e64579b74176060a4459659ec5746c456ef12329e91b4512a7c5","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/final.safetensors","sha256":"75cd6806da0a89ee86f18129a0e6e1e8c7a819aa7298e826a0873ed0b2ed1d22","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"efef3f155519d8aa4b6defe40afcdce702e74f2d76f37b784549b45257d6d3a8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"62b4e268f516db6889a991e3658457259d3b24256b9c8ab0210035775a85dee6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"c47160172585dab1af83c0a06aba3e29c799bbb982b902f55f104870d1cea985","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"03371a0c0bc30bbaf823e256d85219a0af0cddb9977f9bb6116038e7f3c2ced6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"efef3f155519d8aa4b6defe40afcdce702e74f2d76f37b784549b45257d6d3a8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"62b4e268f516db6889a991e3658457259d3b24256b9c8ab0210035775a85dee6","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"c47160172585dab1af83c0a06aba3e29c799bbb982b902f55f104870d1cea985","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"03371a0c0bc30bbaf823e256d85219a0af0cddb9977f9bb6116038e7f3c2ced6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-143.pt","sha256":"5043b17342521641d3ff3c25b58510c33b2cb6416d3611dfabab972d091937ad","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","sha256":"61aa20618fb2c6e1cadd3871e0001285dd2f614b0056ca0021c03a5981129248","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","sha256":"e0e16afce584dcbada260502011818e7a2d4e4588723372b740db6daa4cb927a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","sha256":"988bf544e1c3d92e7dea9a1cc8051747a5d81b7de1f9e8e0e15811cbd0cb7be7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","sha256":"a4ee2e03aac450c5132fe2b905fb84f89c135aea3e6df2ef93dd0bba9269dabd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","sha256":"e877c73f82f5fefd4bda9f3b25f68d10bda6cf69b3f03d5b1a2f1411afc96bea","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","sha256":"83ff6217e245aec96a0b8dfe001866b25a738a108b0b28a6a17485a084efd5c7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","sha256":"d38735c053602e29d3c3d514ca175813c79eabf5429b01072f28a4e8735c8ab2","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","sha256":"cf12ce1da70258b36bda06b690f1ddd982773eefa65f8786caa4e3bebfbfb42f","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"59bab6309eb654b78b2e87b55585869ae1b462916ade568ab974ead6d529b9fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"e3109e5d8d31f31507aefb5dd53a8f6ae000adc4e7eb0f3963b8354c77b4dc2c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"31f79b9c58f8c061796fde2f2ba82ba0a4f62e01e8c43bfab314b8ef51c3268f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"0bb80e28e00923934631b4021787417791d6b5a1698a217af4d9edc30d526754","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"ac1629a3b7ad084bc4ed9a434ba89bb4fc24c3a0fa11adfd24268fcbd0c49a20","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"c01f1634fe5acea342498e39c22966fb24879a5c1e7e6f353fbf28ac9c7efc98","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"ad9a905b0d409d5c74de265d7ff7a01e758f93e4ea7b66a07cee7a6fd43e52fa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"a309e8099a8f89b27b1db55fe5b03edb23e5192ebd660381d9662ea0faf9f05e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"5af606e9fc065f070b1fffeb43d916c8e2d82ade53b1c4adb72cc85c73f005e0","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"56575fac55c41ab70680849bbb224ab5b9e524c881f5354a2e1a6c9e84a66162","bytes":5344}},"requestSha256":"811401bbee02c310f6408e59001e86e2546ea5a349deecb219b5e30e250255d5","metricsSha256":"ac59e89db1f39f51f2a3059f30f56fb680e29228bec5560707e01608021c0fc4","continuationSha256":"73e140a55cbf1d7747659b13087ca8c466a5ed0abfc026de7430f1fce3cad413","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"59bab6309eb654b78b2e87b55585869ae1b462916ade568ab974ead6d529b9fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"e3109e5d8d31f31507aefb5dd53a8f6ae000adc4e7eb0f3963b8354c77b4dc2c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"31f79b9c58f8c061796fde2f2ba82ba0a4f62e01e8c43bfab314b8ef51c3268f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"0bb80e28e00923934631b4021787417791d6b5a1698a217af4d9edc30d526754","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"ac1629a3b7ad084bc4ed9a434ba89bb4fc24c3a0fa11adfd24268fcbd0c49a20","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"c01f1634fe5acea342498e39c22966fb24879a5c1e7e6f353fbf28ac9c7efc98","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"ad9a905b0d409d5c74de265d7ff7a01e758f93e4ea7b66a07cee7a6fd43e52fa","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"a309e8099a8f89b27b1db55fe5b03edb23e5192ebd660381d9662ea0faf9f05e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"5af606e9fc065f070b1fffeb43d916c8e2d82ade53b1c4adb72cc85c73f005e0","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001/continuation.pt","sha256":"ee61fd013965d26af91f4ebabf8b4e61e6881a3bff692283d5c7e06932cbf700","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"78471eb74e19389b4ed65846ca13504be2ef89a61344ed670c7c67c3cbaafa80","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"95613d0a72af7a660027f745623185bf8d778fad5d4067f8ca41a9272fb0001d","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.006-s44","horizon":80,"attempt":1,"job_id":"2481812","recipe_identity":"2307a214ffd389c3293adde39d86c534b84167f84253e55219563c389e8cbabd","loss":3.5002686205334763,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/result.json","sha256":"74aabc88cf7ce1cedc0a2e6cb65734119e98b3554adaeb2f13371ee1fadb9b8d","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.pt","sha256":"1e48226460f1381e948ab67cf6f29fd21998477d1451abce2aad33ea29dbf561","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/final.safetensors","sha256":"97de5af030a78b935afacbb2472f305c7638cb4651d64efe0c0fb8ed2265c888","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"8c21a523429f76b9e42cd4d8f0d0e9eed4674e587e7faf2f744d3812953130ca","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"8adde038c5b8b74a9e55361769c3f11fd080565e8d7b67ce91ee93e598450d7c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"38feeb6caa72d8b1eb420b5e7818a8da5e0f34191b98d98d8548259458648718","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"a217854dd482b81d7640ae51c7dadf4955f7490e6ce6df7ca699ee3948cd876f","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"8c21a523429f76b9e42cd4d8f0d0e9eed4674e587e7faf2f744d3812953130ca","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"8adde038c5b8b74a9e55361769c3f11fd080565e8d7b67ce91ee93e598450d7c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"38feeb6caa72d8b1eb420b5e7818a8da5e0f34191b98d98d8548259458648718","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"a217854dd482b81d7640ae51c7dadf4955f7490e6ce6df7ca699ee3948cd876f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-143.pt","sha256":"d735e182f2b80331f399f627a2e9039265594e189ffdefc20959d900c25bb150","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","sha256":"ec966aff9d4a91e58b7786e81acd09fb5d5d220e186d01fae5f76c1f2c4e7ecb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","sha256":"a6c3a62faad51a88fec3cc7b2f9bffb6052498dd8e48ba404882e629c1de3d92","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","sha256":"b462758be885b237d0a87549ba11708af243e040242f7e9482c4ef06e8b15bac","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","sha256":"dec4258b041430b5c7728f2cdb88f7353c243c22469512ad2d415cc86523ec39","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","sha256":"31e8f27ede017339cfafb34d2a7efb951d4252c3846ce8f6b7572c5437883a74","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","sha256":"b66121a3eb32b991ed8e825b23d5e23d68cd813abfa1a58b02e8cf02e0d69001","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","sha256":"43de87a0786e40294066c40fec30d43a28e99130f0c43c7d39d739f5d8377678","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","sha256":"8215b4e907cb193614f0d443952b4d9a9e318445f2e8f212c3b4a1a46781b4af","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"e3654df1ea4347b64c50fe46936d56ce428498d005c860f00ad244a282ffce7a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"d9ce4eec2bb93d0de875a07df025399ab7032f3204adb711ba36b1569a775ccd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"be4e7b8113b58a68337e83428c00b5fb82ad686029d7c5e99adc370fe1983ce2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"6db1716b7c5eaf35e521f229ae148bd29c2d66259325926f7bf6b7490b50283a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"e9f10e409a9a55f1cfb1631bb5507d83c3e45da4d77204312abd4b925ca5a795","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"12ff3e88fc214d971bd2766e5899a86a0e8d032be59da312c55ae5c6a2082fa3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"3b8d830183853c359375f896664d8c355bd5e75a20d82e3c9f8adec27575e001","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"d61c951313dde5b8b2ddc2e193d587fdb6d4c0c350f89719839f08de43b497d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"0d1374c6b2690a40fe2d870201993e51d555fc4b49e4cbc72573fb9d1f6e5d6e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"88fe6ef3e60e64b34d44fa20299672265525da3caf36ef71d273bf4edc0ce8bb","bytes":5344}},"requestSha256":"75bc4fc94e9d1105240878d347a23f7ac42e49417f992007840c01f7b3558ea2","metricsSha256":"95534941d64bef75c3a433f8adcc0532e10e398e7dc5630a3779e0d326adee78","continuationSha256":"769b74c9368fda24b91e84ac602015d04163b280eb48e877ba52f3502f016e36","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"e3654df1ea4347b64c50fe46936d56ce428498d005c860f00ad244a282ffce7a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"d9ce4eec2bb93d0de875a07df025399ab7032f3204adb711ba36b1569a775ccd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"be4e7b8113b58a68337e83428c00b5fb82ad686029d7c5e99adc370fe1983ce2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"6db1716b7c5eaf35e521f229ae148bd29c2d66259325926f7bf6b7490b50283a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"e9f10e409a9a55f1cfb1631bb5507d83c3e45da4d77204312abd4b925ca5a795","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"12ff3e88fc214d971bd2766e5899a86a0e8d032be59da312c55ae5c6a2082fa3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"3b8d830183853c359375f896664d8c355bd5e75a20d82e3c9f8adec27575e001","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"d61c951313dde5b8b2ddc2e193d587fdb6d4c0c350f89719839f08de43b497d7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"0d1374c6b2690a40fe2d870201993e51d555fc4b49e4cbc72573fb9d1f6e5d6e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001/continuation.pt","sha256":"d70b9f941040a9fdd49cb1648d77d099d74752d98a72ad160e9c0df04403694c","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3abcb8eb5b42802b8fbd9f214a14d9e4633920aeb824cde591a1e21958a66381","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2307a214ffd389c3293adde39d86c534b84167f84253e55219563c389e8cbabd","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2481813","recipe_identity":"49c74d8b050423808b0da4cc32b6ade41e0764e61dc6ded47f1f173510058c6f","loss":3.4847878325903787,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/result.json","sha256":"d918320e347b89f492f7bbb6123256dd8ed8dc835e595f88ccec99949c75fae2","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.pt","sha256":"1525ba15f2eff5c7e260a544e084f5bfc0ed6bdace307184b5a2d92cf73a06e7","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/final.safetensors","sha256":"fbecf94796945704989d4397df5df7ae13ada55d7a1e88fe58aa1de28e0ffea6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"6923d4b5d7540416b3ee1663a41cab8fb397eb49c25742991add0db451106282","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"91bd1fc08f1549df05c64e3e4a312cd371d87fb29af8b5e343da43e86442fa90","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"9eabead219b8054002d02812c6cd164aeb887c982fd6f283698887d93ff2d157","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"e6d3ec65454f22a0f23c05db525680ea9233519db9580f21f455ca1c1bf8afcb","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"6923d4b5d7540416b3ee1663a41cab8fb397eb49c25742991add0db451106282","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"91bd1fc08f1549df05c64e3e4a312cd371d87fb29af8b5e343da43e86442fa90","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"9eabead219b8054002d02812c6cd164aeb887c982fd6f283698887d93ff2d157","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"e6d3ec65454f22a0f23c05db525680ea9233519db9580f21f455ca1c1bf8afcb","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"cec0b21744fb2f2ec3574a1971ea2849fc3d790e7462e532ef50782f6a6ec474","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","sha256":"6fc8f66b59a8ab3d14bb86a850818221e47a6c1bac5a2de07a73b1fa8a4a2c56","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","sha256":"4664a07f374f61590ee7778db0635cf1fee38de910816f22d1c7af28e78020d5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","sha256":"0d2698bc0010fb4d0a69a3aed72a5082d86bd67bf2d3716511fa3d3cb38318b9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","sha256":"99eab247ad20bb00820677a4dbdefaac8c53ba7a459cf7c4f2fcaf42a0fa09df","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","sha256":"3c0e3818f16de810735ba6545c76f98a05aea7eb0b5e8bfb5c1705e4c0e84627","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","sha256":"4983929f4581bc3e778db96a8a1f6f8e077449b4aaf9c1bd40ad80080b3a7ea5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","sha256":"2b3fb4be5c07b5ae67fe97ea7aa6533eb567173fc8aee5ae0d79d00072e9c522","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","sha256":"0bcb9e8b39f26d8d40b4fd38127a61a924be1b2b06abd678e81c9360d998137a","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"fecec5b9a9e6e5a54146da904ad1987160a1980b0371dbde855bbbbc1fac7d0f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"8b1f198c7b31b4c723f122147425b3d4ef29c4782f7c96c3bcd71d1791a03085","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"24c631d257c4266b3b72258771dd1d639914a2f6135bb0dde4de53ba2c97dc76","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"2bf6190c8c0aa085d450cb2dcf0cb8e0615d5142accfd9138b7293eb1e75aead","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"04fa823295473125cf626ec8cb5cf6a89b3bb22fd38ebccc9b537ebcaa856fe5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"582093db9f4646b8aa35394eae927c1573d3d7c37218996a05eaefa3af206df4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"14350a8da92951f9aedd8384d29b534d418bb5537ba1fcd7a159aa22476ffc0e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"c6181b7ff766816e19bc865f85ed63fcf3d2127fcc9eddda972328c9a31d2560","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"da708b7d419115ffba81a28a64ab14193ff9bc74d7c0068f1d9a13e613f70999","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"ee9d77aefa53f9f47a8dbc9c786ce74d905b63f8b1c89125ebeda30771d4d37a","bytes":5344}},"requestSha256":"6175b2e4781e076b80a310e1d8020bd4583fcae8b2d378bd1a09d9f2f8d9371f","metricsSha256":"08b08502fda6a76d9fa1f5807b16271e06a341cf1fc6f294f613590f60dd01b4","continuationSha256":"8f76be896057167c052a74d746aa76e49357baa84399cc7852d950674aaeacb7","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"fecec5b9a9e6e5a54146da904ad1987160a1980b0371dbde855bbbbc1fac7d0f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"8b1f198c7b31b4c723f122147425b3d4ef29c4782f7c96c3bcd71d1791a03085","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"24c631d257c4266b3b72258771dd1d639914a2f6135bb0dde4de53ba2c97dc76","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"2bf6190c8c0aa085d450cb2dcf0cb8e0615d5142accfd9138b7293eb1e75aead","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"04fa823295473125cf626ec8cb5cf6a89b3bb22fd38ebccc9b537ebcaa856fe5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"582093db9f4646b8aa35394eae927c1573d3d7c37218996a05eaefa3af206df4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"14350a8da92951f9aedd8384d29b534d418bb5537ba1fcd7a159aa22476ffc0e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"c6181b7ff766816e19bc865f85ed63fcf3d2127fcc9eddda972328c9a31d2560","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"da708b7d419115ffba81a28a64ab14193ff9bc74d7c0068f1d9a13e613f70999","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001/continuation.pt","sha256":"7d38d5217eb09a63bc75f5cf6c5a13c2f704179a65e6fb465d4adfae6ab53fc2","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c62f1e9c3aa5c978aa3f4b969c5481a5cc3e1c89fb29d640991b936d1cfff9b4","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"49c74d8b050423808b0da4cc32b6ade41e0764e61dc6ded47f1f173510058c6f","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2481814","recipe_identity":"ee774196f9aa83fc3e112fd757c98e96795ed4b3505214539822e1c038a02460","loss":3.49333101963653,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/result.json","sha256":"a1d913825b1971321368bbf56839d58645636f787b948587bcc6661d5d876022","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.pt","sha256":"092416d53f4ec5cfad1d451a70c57e69e92b46c2584a733f7c918cf4e34bb4cb","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/final.safetensors","sha256":"6fde1138d3883f7320896cc1b8ab4393f8b298ac669a575068cd39ca16d3fd58","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"579ce1137796f5bcf97915ada0bf40dc398575448ae7466f9bcf82dac5fd0087","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"353a4fb13284e30dbe85c81d5a567de43ffbe9f5e85184a6f38252419972815d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"cb84df51a46176b2f4ab51d2422f7ccc3feab449755f2d57fc9d1bcae3427cd3","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"6a11aae5b575bd9d4cf43733990d57465dee16b5255a5297b2a5e86dfea42803","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"579ce1137796f5bcf97915ada0bf40dc398575448ae7466f9bcf82dac5fd0087","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"353a4fb13284e30dbe85c81d5a567de43ffbe9f5e85184a6f38252419972815d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"cb84df51a46176b2f4ab51d2422f7ccc3feab449755f2d57fc9d1bcae3427cd3","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"6a11aae5b575bd9d4cf43733990d57465dee16b5255a5297b2a5e86dfea42803","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"6399104e38bdb4a858a21cad73be655d13d7af8a11eac1ac6bc272d27cba0b5c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","sha256":"294891106f18fd2745aac8ee41183306113e2cb630e85b2940ed13ecdbbe8666","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","sha256":"7d21418becae7c2d253b16f076954705c9b74770540dad765d88c3571bdd810c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","sha256":"1be3396f1ae192ebc0c9a08d5b2bce5fe302e1da3d14931693f62ef62fd31ec6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","sha256":"8421132fa41b8c2dd1af9820a757e31445312ec37cd7a888d377c6c37ce29566","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","sha256":"02708a71b2b7d43785c8cb69c27b81185c22d26388cc5a7cb7dc22a52ea5923e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","sha256":"c67520c217e36fc83bcea9d16c489182b11d6b45154c6370bcbb2691e8304753","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","sha256":"3938579d9da5438d5992224d9883c94b3027318021de9ba38ea109c8a6972338","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","sha256":"6076bb12f1c041807129adefcd8bc487e4b8bfcf6459f7c7ba197f9de19a6f66","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"6711666ba5e3ebd81526ba9fb81a11ae2faef380a9c82bfea86d46e9b94d1b71","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"f591e9e6f06d06a7ec8fec54f33ff6325f340525ae6ba7b3a9b4debbb8182927","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"2314d9f426124caf5d47fbb62a8962bea12013ce1d36f2034b99962f3edac51c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"9364090f275965a02c664c1b0893ad7ffc7440d1b314928f5c257f4c9f77cfe1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"1b0861df0d250a4ca4b987dbc0fead7a034cbe5269596e8f14c8ffd171d35cc4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"d7bcfda9b3eafb89d233088e7829a566e353979410ac911d3b03b4e93a0a9042","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"fc9f15bf8c5d6d12bac87ca30dfa91d1f7c8391607fb39b7d2b9531cc2f59732","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"f10f43bf47a49304fe68fe0dfe4aaa693b69d395228e16436325320103a0ffa1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"07541bc765950faa5407086098e142ec26c5574b705018c7d301d38dbcd60c17","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"6bf3c418df9dc747d9a06946474c3e8e83020a5883a3ec200dbffa7080328dda","bytes":5315}},"requestSha256":"45069a66096fc16cf8bf3751b9400a416c66eaf0fb29470d4ec78f39b678a367","metricsSha256":"8df559b4ce3c802c5b3c8688ec526a46db48045cafd976c4b100f5df41073fea","continuationSha256":"4ecc18d136680ebed64326d8e0fa5b031b55259738e1855e18eb04dac6b28837","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"6711666ba5e3ebd81526ba9fb81a11ae2faef380a9c82bfea86d46e9b94d1b71","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"f591e9e6f06d06a7ec8fec54f33ff6325f340525ae6ba7b3a9b4debbb8182927","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"2314d9f426124caf5d47fbb62a8962bea12013ce1d36f2034b99962f3edac51c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"9364090f275965a02c664c1b0893ad7ffc7440d1b314928f5c257f4c9f77cfe1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"1b0861df0d250a4ca4b987dbc0fead7a034cbe5269596e8f14c8ffd171d35cc4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"d7bcfda9b3eafb89d233088e7829a566e353979410ac911d3b03b4e93a0a9042","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"fc9f15bf8c5d6d12bac87ca30dfa91d1f7c8391607fb39b7d2b9531cc2f59732","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"f10f43bf47a49304fe68fe0dfe4aaa693b69d395228e16436325320103a0ffa1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"07541bc765950faa5407086098e142ec26c5574b705018c7d301d38dbcd60c17","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001/continuation.pt","sha256":"f9ba5426c15a04695e63dacf7a1b745f8262af0751d99e581ba0ce33cf976ea5","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b201b665385ce6b61f50f6a3280d284d30af2d6c1b37d19a25221cb2082eef38","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ee774196f9aa83fc3e112fd757c98e96795ed4b3505214539822e1c038a02460","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2481815","recipe_identity":"0479129ece57f04975245211d6d532fc5b04db4d36cdaf128f7699b8e173014f","loss":3.493220159276739,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/result.json","sha256":"fa9e6e86854cee100155011d72abdb4b96bcf1f30d23cfd50885e165d37e3aa3","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.pt","sha256":"45df1872bc188659dc24b29b97b6a47131dbb02c4db36992b56d1db810cc0d1d","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/final.safetensors","sha256":"a517788c5eb19110dbac1151cc1068803f3f5a8c1e77384968b9a290e47554cc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"8d7bbfc1b4a7e82f504961785b67bc32095bcd99b13e0862ea443e8b47affc39","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"0619ce40b2be5f1b11c88552959e8c4574b0f47025829d2533b34f5afaf4d556","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"77c1d944903dd486845074996f734f0991882e17fe7ab038ab6748cc31a4002f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"05aad7a9602187a41a294ae6739a5a9d6a88569f9dc96d414e7b18c88912406d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"8d7bbfc1b4a7e82f504961785b67bc32095bcd99b13e0862ea443e8b47affc39","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"0619ce40b2be5f1b11c88552959e8c4574b0f47025829d2533b34f5afaf4d556","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"77c1d944903dd486845074996f734f0991882e17fe7ab038ab6748cc31a4002f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"05aad7a9602187a41a294ae6739a5a9d6a88569f9dc96d414e7b18c88912406d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"8b13b0b787bbda3ceb33f632a890659cb189ebbe4361b4fcc7c64f96555a9bd8","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","sha256":"d790af36e4ad1c6857e20d1b3a83a8efafe8342e5366b5163ce968d941348ada","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","sha256":"f5ec5e715e09886d190921d1185443af2fe5308da0abab968720a9b90713315c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","sha256":"75d9457f39b3cad95b5c8b3c7253e1d6c0e22da6071016a5e4d68d4757cb5664","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","sha256":"c5d2a618ce2f0a2fbbb0564f6bfbf0629503e7a84a32e66543f6d00d29a59ac6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","sha256":"8e5e0015558df236be1bec0d06c768a51b84fcaf3e68cb14530573b835c8eae8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","sha256":"002d0f8fc1ce8b435a721661a1ed0fd47418823ae54fa8416462a5ab569adb1b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","sha256":"be7329cb7ef23b4c1cba491b6bf1a521de6f7c579670c2becc7949d789ab2d47","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","sha256":"0bd38b51186e8a9368ed372eacbfb8cc4aeb9003b2a4bb6a765b43448c0c7036","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"02484b11ba09fc2082d01402293e903982a3ec9c4a00a7e3e98bc3b2c6b903fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"538195540a03b59cb2ad31294eb4949e0fe71b3ed3ec6766510d0315663b765d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"c79f1db301e9d0c4e76f9867e09f6e4f050f25206c204a57e5dec9de53748d41","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"371028704a915c360652d9f610cb85afe84e3ecbd3254277008d2ed23ef9c38a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"90d7de16add224fb81f8011a9c19da78c33844f9a7b648a01598aa47640abe85","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"bf0305feb63b4a5e96128d437c5d38b7f0fdf41314ce8b3ad79942a7b9707de3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"b63687a85ad111bf8f03f54aba909c918bc8fe160669e95bde6bdaed90b5e293","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"46c3de7dcd38399eeb39d91b1cc6f3c2c20e61ef874ab07a44f8f579a8609671","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"b10df18dc4d579fb2c81d335dab0cf73d3b42ae39db2bc4dc76928d382275567","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"c7c19facdbab68f3e9f732286cf1db0185b3cdcbad5f19313171fffac51ba934","bytes":5344}},"requestSha256":"ee46cf6ecefdbf7488242ad5a0f3a0c5b057d238a2db315f1394087d015062e6","metricsSha256":"8c0d76e0d85ab20304fa459d2c55f7842c6866afba821ba5c5b89918d757df4d","continuationSha256":"8e6761abb3e498ecf39a6a1ec256d771c3617925b097bff5f4410caa245f433b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"02484b11ba09fc2082d01402293e903982a3ec9c4a00a7e3e98bc3b2c6b903fa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"538195540a03b59cb2ad31294eb4949e0fe71b3ed3ec6766510d0315663b765d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"c79f1db301e9d0c4e76f9867e09f6e4f050f25206c204a57e5dec9de53748d41","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"371028704a915c360652d9f610cb85afe84e3ecbd3254277008d2ed23ef9c38a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"90d7de16add224fb81f8011a9c19da78c33844f9a7b648a01598aa47640abe85","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"bf0305feb63b4a5e96128d437c5d38b7f0fdf41314ce8b3ad79942a7b9707de3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"b63687a85ad111bf8f03f54aba909c918bc8fe160669e95bde6bdaed90b5e293","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"46c3de7dcd38399eeb39d91b1cc6f3c2c20e61ef874ab07a44f8f579a8609671","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"b10df18dc4d579fb2c81d335dab0cf73d3b42ae39db2bc4dc76928d382275567","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001/continuation.pt","sha256":"9947fc81e36067d36ea200b473699e85f5534eaf5313ba769062796c559b198b","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"05e108a2ce9345a4ae2ba41eff7a1d08bc4e76a41d7f192204706bf074b757ab","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0479129ece57f04975245211d6d532fc5b04db4d36cdaf128f7699b8e173014f","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2481816","recipe_identity":"aa4302c20a66f16330f92efe19f6fd6bb037b2e17f79e2ad53ced1b0932ff01b","loss":3.4940040759130357,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"fde2bfb6b02d09e13a8017144cbd665247156af9f7fb2db38d8784509c672ffc","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"5507d59fc00020fb6fdbada80d6ba1e4bd0711b20b416e63c73de205d44fe0de","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"a702fe23b129ed0375c92f335ba4c50cfd4ae339b5b991c3625d6876acbf0042","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"7574484e0f48e22f580a6cd64beaf32018fe8d5cc13f9f95cdedc64697b7a453","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"2769a280ab70531d2bdb625c3bf50acf4a9a40f638520408b9f6a2c4bae31d16","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"92d19aae10748bf05656db7e320fff0715747b184ede1acf864ceab2e0af89e0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"9f6951988be64d338e6e7ca00aaccb01644e47db3d176a3d44a586653b60d6a8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"7574484e0f48e22f580a6cd64beaf32018fe8d5cc13f9f95cdedc64697b7a453","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"2769a280ab70531d2bdb625c3bf50acf4a9a40f638520408b9f6a2c4bae31d16","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"92d19aae10748bf05656db7e320fff0715747b184ede1acf864ceab2e0af89e0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"9f6951988be64d338e6e7ca00aaccb01644e47db3d176a3d44a586653b60d6a8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"f29d6e3681247bbbea1e3c61dfaeb1e10a560ef05b0cc4299c7fb50291f463ce","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"db3e31716679e92fe4a27c3c9b7e6e6dffa268d5b1c57db80fd149fbecdc4a6a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"d088bcfb5e740f2f2da7388fcc9e07ced11534a6e4cb6e370934cea33f6bacdf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"055222ace1a74f5c18fe1401a5a5e41c76914c32f5fe68e740c3c8d2cbbda5cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"fd69bcc74fe692ead1e30d46aee4d50067d56e63c567211f27b4e5efdab248f5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"4ba4e47d8bf4fd5bee34fca2f87dbdb441b21e74587e34cca0a7f9fd60a820d8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"88839671ce2fa09d5d9f37b26974a02161937fdd739e14637856b60378d57639","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"ec32dbd1374af11a9de993e1ccba73f4b8ef6a95f9bf7692ef56fea2ee37174e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"544a21a7eb68b1a9c40b96dae09515bbb007905363be73dfac04bdd833d53749","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"b06036928109d4882768e9db3a7155cf3011cc0f87c3ef0eefab0b70069743b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"2ea3b6ee13994d5c7c7451a64799dfea8aa9f4630c8f9826e50e43e9606cd88a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"20dec9f92e396c3f2289bc463781fcc0c2cd0ccabb7884d90de76343bb3c99b2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"30c82c91e16d5b8b091fe877322cacb526d6a652fefe807d36d15dcddf08398a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"da982ff5e0ef32f03e3cc4614b3f0aa7f1aacf32e04943ee222c0b823c3fcae0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"703ebf33d9bc931a8a11fe80e982427514da24409e767a5243cdf46d241b2456","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"267c76a7fac155d86eadd13209dfe642804814bd94c6cc21b9652bc967eb04af","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"c8f626778e5ce70837df65755d62fd507828ef591913a139bd03f4fb72915a4c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"e8bfaffe2c146fd46e5fcda34ca47536c677a9aa01339bd0f6924933cd969e9e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"e5219ecbefe42252fa50fe5db48f36ceea9af8ec413f8a719111c4e941ee5b34","bytes":5344}},"requestSha256":"987ff537fdb3565ea69c01a5461b23b63f8ff225961d6f0223d6b2737e455380","metricsSha256":"5682163dcec29cbed9658ca0390bad8e10ef2dbd86f48c6e2a03ee08c93f94c4","continuationSha256":"0ae81e1c37f696a5d449bc1795eedeb511a3b7d67c1ecce07e3509b00ed787cf","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"b06036928109d4882768e9db3a7155cf3011cc0f87c3ef0eefab0b70069743b0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"2ea3b6ee13994d5c7c7451a64799dfea8aa9f4630c8f9826e50e43e9606cd88a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"20dec9f92e396c3f2289bc463781fcc0c2cd0ccabb7884d90de76343bb3c99b2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"30c82c91e16d5b8b091fe877322cacb526d6a652fefe807d36d15dcddf08398a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"da982ff5e0ef32f03e3cc4614b3f0aa7f1aacf32e04943ee222c0b823c3fcae0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"703ebf33d9bc931a8a11fe80e982427514da24409e767a5243cdf46d241b2456","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"267c76a7fac155d86eadd13209dfe642804814bd94c6cc21b9652bc967eb04af","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"c8f626778e5ce70837df65755d62fd507828ef591913a139bd03f4fb72915a4c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"e8bfaffe2c146fd46e5fcda34ca47536c677a9aa01339bd0f6924933cd969e9e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"c72ea108bcab911b0376cb6c28ad106e51995f5c11e6f8b8c408ba5e5ab2af29","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8d720a3e981bd355b6f1f21443093cde8b3217905f5cbce3935c8d946cc2aa06","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"aa4302c20a66f16330f92efe19f6fd6bb037b2e17f79e2ad53ced1b0932ff01b","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2481817","recipe_identity":"5e11520544593a2addcb4241d09a18ac6602a2bf5d87ae824ed9a8c31ae58c0c","loss":3.494852123273075,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"1e8499dabcf9a68799026040dfbbbd1f23659a46651b06e596b42330ef2e3296","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"973ae73637e62d12c899114de306e39a67e3e1878994fa0a33bb3014778c4910","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"738590317a7ec2476fae0f4917f2eef0bef3d4a07a7c586616875d191776f6df","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"fefd7e6b4475f2cd904dbcdee4606a91848ab17e87614c9103f20a2c50f8a9f7","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"2cd560c29777da9d5b33f27362b5be17632a369dfb5cdcbbe63c48a1f3e74e25","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"2914c06c910ee27e819996304777a24ec7f6c3ee0ee55b0c7632c9ff262f1358","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"70a7dd86f1416a0746332114229c9f3fa4b791573bdce15e57a7889de2bc0475","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"fefd7e6b4475f2cd904dbcdee4606a91848ab17e87614c9103f20a2c50f8a9f7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"2cd560c29777da9d5b33f27362b5be17632a369dfb5cdcbbe63c48a1f3e74e25","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"2914c06c910ee27e819996304777a24ec7f6c3ee0ee55b0c7632c9ff262f1358","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"70a7dd86f1416a0746332114229c9f3fa4b791573bdce15e57a7889de2bc0475","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"97a2636ed288b3ff89ee278f531d8b9f474459eea7a05ea813dc1a70e909d616","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e8437defc7f3a6edf1ecbebb1ccd4577fb655f94ab587b4671b5999f7e09ed2d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"7056a2a55a5e7319cca26b93c64ba1959b4baf96685b329b8ac87c4c521785dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"efe4432ce0f0b7ee07dee4ef73f8e22afb98b6706e4c4adcbfbdce22bd61a815","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"fed7a39fa6a77ca743da8e87983e4817842fb0d063bca69d5c70e13697cd801c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"dd2be832e8a3f0227bd52d3790764369cf8ce8261b74369e62a730005130c3b0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"98842232e4429ab708457aa646f3cbc8aa46ef170cda6fcd7119a0916dadd23d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"0c792d5674348ad3209c2316f1d8d6c1b5c95a196d4712dd2d3be03c0c66da6a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"6dfac06ceb2f274e37a37428a99cfb1dcbfdc072c0bb5af1d2e577702b52f3fe","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e0b7d6e394867f69a1c7f9aaad2e824414d422282daeee03c692f07004931064","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"98645167fbc84f5e1813880635bd17d5237995e5b8d57b644d4ce7dfd57818c6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"ebc0d4ef6a111d6f00d5c0e79bfdc42e7d17d34a118e179f90fa652282c9ec59","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"3e1ef8933d9918992fbf9bc7e913d6d386475f8b3bd649841beb0ca835f62f2c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"d8851a232768e6ec9e2baa3043e8f130b3f4ae21d3ca40c1101d7f21322edf4b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"096012e2e6c3ef8d82394eb327a72ccf0a1a66f0dae377abd6036606446d6e50","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"f7d0ca36dc51f2b108edb123b5359ff8f7a5d7d927564a839a693be52ac95e56","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"f8bd75dea5303d53e62996d401b03eaaffb505f797a5c78a35a9371ccfd38afe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"7c7e79dd079a603a0922df5d4b5fde0ed12ca455c0133beaf564aca8ad950bb9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"8380beeff6d7bd08462f94c6ba7b07350deabea6cc52870793f489c8e37461d8","bytes":5313}},"requestSha256":"ad129af63d84c512b8b7ff0a28b57a0ed1683ccea8e47d979440b128e84a9221","metricsSha256":"324b8bd191cd142f3b55744981e6ef30af448134db15109fdeb86b1841b15568","continuationSha256":"0aad7fb1a2d3d575f4a1b9aeb79f4c54717ce831b88b9f70fb71bad87da1af30","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"e0b7d6e394867f69a1c7f9aaad2e824414d422282daeee03c692f07004931064","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"98645167fbc84f5e1813880635bd17d5237995e5b8d57b644d4ce7dfd57818c6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"ebc0d4ef6a111d6f00d5c0e79bfdc42e7d17d34a118e179f90fa652282c9ec59","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"3e1ef8933d9918992fbf9bc7e913d6d386475f8b3bd649841beb0ca835f62f2c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"d8851a232768e6ec9e2baa3043e8f130b3f4ae21d3ca40c1101d7f21322edf4b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"096012e2e6c3ef8d82394eb327a72ccf0a1a66f0dae377abd6036606446d6e50","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"f7d0ca36dc51f2b108edb123b5359ff8f7a5d7d927564a839a693be52ac95e56","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"f8bd75dea5303d53e62996d401b03eaaffb505f797a5c78a35a9371ccfd38afe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"7c7e79dd079a603a0922df5d4b5fde0ed12ca455c0133beaf564aca8ad950bb9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"8756b84e30139fd2dbc0a2c839e31778b79e39b6859f0be7d7a66433db0bf177","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a3088b8abb561806032d0f55e8cd6d6e6cd56b0566940ad7dbadd49b9551050e","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5e11520544593a2addcb4241d09a18ac6602a2bf5d87ae824ed9a8c31ae58c0c","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2481818","recipe_identity":"9bf3b2e4247426c07cf9670d2da942ed6a66266f2e0b0699b70ed4c2e93e7c52","loss":3.4974370610999816,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"6b8c432b9accfb60f193951b7c1471df294a6a14f027599f05cfc26fd6dfd24e","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"1dcdb7b21e8b225ed531e8d2d86158ca0d040d5e011c0d36ba6efb82c75ba202","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"840e033e7c638a86a3c64e28de9f847d53415e56fd3283aacea0f7bcb549b43a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"c19dec59745eede6fce43cacdb7f05286e65426644527347ef74a1fc03e4ca1c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"d88d116948070a5d9c59c25a69854f1c1c6fd40a28c472d87da2f10585df8738","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"abf3ef05f2e5a4767ca95b4be207f6aac2a5f0f4e3db5b3d1b4f5b0f6dec0abf","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"17420894f0f51cd81ff57557cf0ef29493264a577ac067ef2e626c0df60af35d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"c19dec59745eede6fce43cacdb7f05286e65426644527347ef74a1fc03e4ca1c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"d88d116948070a5d9c59c25a69854f1c1c6fd40a28c472d87da2f10585df8738","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"abf3ef05f2e5a4767ca95b4be207f6aac2a5f0f4e3db5b3d1b4f5b0f6dec0abf","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"17420894f0f51cd81ff57557cf0ef29493264a577ac067ef2e626c0df60af35d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"a82b6c589d96dcb477c3900f42537fad4ecc6d373c6f2f53d89a5cccef1cbf76","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"c0c3228e7750b9a644c22e0546814b251aa6cc05df19fd74b02759f255908f25","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"5c8ccb80eee104bf0057728a16a26a09e18f143d88276ca51935326640889eb1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"ee240c5cdabfa2982d7d86491adbfe53a128444d3610882a0de0d8af6f652aa0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"8c1ef850ca0c49d59a5da969c467fd609075ea5bdc83961e339cf605504e9917","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"933ff467b2718dae7298fb58e2e0eeddd0de07f1634bf16bd726917eda03cffc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"a8b3c1b6bf6b2242b5d7168862622f6b312680e1a669f536a5e584ff78408901","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"af09672c7db4e9b298907bdc707a417c883353c6cfdddc29e7a6b57e25a5cd83","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"5f2f4f12c5842d27de601d0997a4a10b0a10392a6b503dd51d69ce4d66bb8b86","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e7d0de7d328d711233ac451477a9c8dbe5a709c0733f597315264b25a6cf06ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"29ce46035055b284b19c6bae8d8355ca734d3e3670f36987b2d426cb05e8d61a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"abe1567e0ec94bf21597c5cd694cd2154af7e1a1812b564bdc29fb8ed7c8dd3c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"658cefbdc3d5f49ce742d62172cc1f5812942dc5b4583aec9b6883973b71e38f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"5f2e8a926154803b800b5e9a055afafbdd95c2759a84494f94d2ddf042e0c401","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"756d6a8498dbde29284487c72461f281c84432f53f217e1bd989da9c824f4317","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"28f2246fc5281e310c51ef9850de2ee0f86ed06665dec1dda3515e37aaa75c65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"e07d2585e1b382df0761ec785cbb0c5b83557ae85bed95d2ededaf8dbbd4c27b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"c6af6edae6fa3df0136fd89b3b73b2a5323643ebabce993cc903a9bc18cabd61","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"bcedeaae57507795d79484e1458407237bfb99957dc70a539db1d5a632bc19a9","bytes":5342}},"requestSha256":"90e7bba72a63431c254673abf2abb9302985ff53cd301c9d9689b4b397cdc907","metricsSha256":"bb687d979db7015860dc32a26c2f78fcd6b04e389c917c407c94334d3cb5dd30","continuationSha256":"a0e7b9763509a2bb5ec3687f6bf0be556a3b4bef9f55a06d63f4683f3ab8c12b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e7d0de7d328d711233ac451477a9c8dbe5a709c0733f597315264b25a6cf06ba","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"29ce46035055b284b19c6bae8d8355ca734d3e3670f36987b2d426cb05e8d61a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"abe1567e0ec94bf21597c5cd694cd2154af7e1a1812b564bdc29fb8ed7c8dd3c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"658cefbdc3d5f49ce742d62172cc1f5812942dc5b4583aec9b6883973b71e38f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"5f2e8a926154803b800b5e9a055afafbdd95c2759a84494f94d2ddf042e0c401","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"756d6a8498dbde29284487c72461f281c84432f53f217e1bd989da9c824f4317","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"28f2246fc5281e310c51ef9850de2ee0f86ed06665dec1dda3515e37aaa75c65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"e07d2585e1b382df0761ec785cbb0c5b83557ae85bed95d2ededaf8dbbd4c27b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"c6af6edae6fa3df0136fd89b3b73b2a5323643ebabce993cc903a9bc18cabd61","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"aca946fb66419202f3fbedbd32ff69cfe0f839feb33580f47e4b9035619f891d","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2a60dc08718f51d6341882510ad4d120ca781d12ee4c3c118c3d3f061bc9eb48","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9bf3b2e4247426c07cf9670d2da942ed6a66266f2e0b0699b70ed4c2e93e7c52","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2481819","recipe_identity":"19e59a263a3cf218909883df36532d62341c3c0c35d73514b93c834483a52898","loss":3.5042184075782337,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"bfd59989e3623b5e413c66e79bdcbfe2095853f1a979089f35b6efa51a9841aa","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"399509a5fc27ad1fc86129c7ce1405a78a2d92b4e570083dc91977df1b79c1f4","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"2f11cd681d0dc9261a38f11ec2d104d7dd5e6a055d96c36402da08042481dda8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"34b6947eac8b027a48e0580ce964e21362ad1fe9405a9499fd255a52047b987a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"cc9aa8a6f4f412b238fc182079e6f35f3de7c46efe2e9864fa2865f7e6591191","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"d85520473b3f6dbb59572c3549738eaaf9726e28ab2c1c8c0d4009d4c493c3f5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"27f2c8ef1163ed80be2fdadd18ef29c1bee35096d828c015d087d8caecc999d6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"34b6947eac8b027a48e0580ce964e21362ad1fe9405a9499fd255a52047b987a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"cc9aa8a6f4f412b238fc182079e6f35f3de7c46efe2e9864fa2865f7e6591191","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"d85520473b3f6dbb59572c3549738eaaf9726e28ab2c1c8c0d4009d4c493c3f5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"27f2c8ef1163ed80be2fdadd18ef29c1bee35096d828c015d087d8caecc999d6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"ca1423e760df68db29f94743b8aceaf9b8ba2538e054db6a35f53b32a81b1931","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"55fbbf3aaf51b7ed22ce8756db85d280093d3d9ff674019d7a3daab7bb3b8dd1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"a77201c61d2cb64402f76a3182013d6e64eeb64dc9040b0f7bf0bfbf7c0ebcf5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"9833e0e540a19075b4e5e991c6ad3dd1a3699007568e6840e6ef15dfe19ca04f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"856bbd62663c9ca8aa56bfe83938bb1f7522afa4192641fbd26cba9426071123","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"49d814536cd9fe5573c3e923e3152450cfb88c86403fa06938c427658b52413c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"604a165e1c9f2a9ad9d2459ac361311c73842cfd2b90b7f2f5666c691249ebb4","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"f5dc6095e432aa29c47e82603cb9564458465c53d674aaccd22ccaec43a07e71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"9e927bdcb6eb5957fcbaccdb025e77f32fbe0b9aa69f0a67324e1cbb752ddc47","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"1f0c2adea6a68f992f222c24fc24e09258b3a4c38fc25b2c1832007a4ddd1a7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"c22cce35d1aa5a280f49c7fbccff57c19a11550abbcaf5a015cc2ce1f1362a94","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"2a5651d3d73e59bbed2f39fb819ca738bab2b66568300775b0c54ae8fceb8c41","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"be9a1b41bb9738e6d7b77a7c346b1e21db64e9c1971353e264e87315e93bfd32","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"8ec473fb694a8f3671779f98111d83bf90b30e6f11338e48e1f93809a751d9fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"d62bbba2b31be32a8b645eec8ae9c93581a2c8a0e392295466c24db2a478b35b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"fcb7654ea02918158c53d8bfba340f95b27c649c205da9a5bb2cf2b8e607a2d0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"211988b166fe5ef30fd6ed3f5b006f54c8d69a992d809a34b17973fcbc851b9c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"bac3f2ff77c383c6d0af49a706d2eab6053f823bd8f30fb08396e4103a5ec671","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"6e6e2dda54662900b92d9703fa57913271edcdaf1e3d55074e0cce2cebcb3c9e","bytes":5257}},"requestSha256":"827a940d04a0fa48c1c345e8462cdbde761ef6e183a8bfa793741373d2135f11","metricsSha256":"6caff6bc900c985b36dbba9bc4a5994944e257b9ff2c207afad4f5fddc0db70a","continuationSha256":"d81c2f6f046284115ff23412d35a8bbd499b6f4dd5a782a0d88701dd13982304","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"1f0c2adea6a68f992f222c24fc24e09258b3a4c38fc25b2c1832007a4ddd1a7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"c22cce35d1aa5a280f49c7fbccff57c19a11550abbcaf5a015cc2ce1f1362a94","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"2a5651d3d73e59bbed2f39fb819ca738bab2b66568300775b0c54ae8fceb8c41","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"be9a1b41bb9738e6d7b77a7c346b1e21db64e9c1971353e264e87315e93bfd32","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"8ec473fb694a8f3671779f98111d83bf90b30e6f11338e48e1f93809a751d9fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"d62bbba2b31be32a8b645eec8ae9c93581a2c8a0e392295466c24db2a478b35b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"fcb7654ea02918158c53d8bfba340f95b27c649c205da9a5bb2cf2b8e607a2d0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"211988b166fe5ef30fd6ed3f5b006f54c8d69a992d809a34b17973fcbc851b9c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"bac3f2ff77c383c6d0af49a706d2eab6053f823bd8f30fb08396e4103a5ec671","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"bcbb0bbdb0c678eaf43bcbfd186be364949d96c23f236decb765882c9e2129e1","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0695af5503ef72ac20eb9d69187842c955baa5331ed727f16df81c59243e5c54","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"19e59a263a3cf218909883df36532d62341c3c0c35d73514b93c834483a52898","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2481820","recipe_identity":"31081c272aee34a76083bf9e7bd699f7687fa51b1bd59f696bf40303962ed24d","loss":3.508940097905032,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"31c74999357a09545a581fd758b6a1dc4fd36a500d971d3309debfc7ae39656e","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"5f25d376e10c1e875454b13467c79b89227a35f499eff46d8d613fe0a48418f6","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"bc3be9f55788df036f7493ae239cd1ca52ae90aeeabe490188a849c0a0fc4693","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"5a4b3d4eac0e44abeabbb7303465cada2f191b4fd58367407378683fd95931a1","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"98a0d9471d7ee36ac6f74d697881bb467bda26dcc2a07b5e4e5aa1cb48ffdbf8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"56c3f116152bd5dd351072172ec671849bd3ab26406b774cf7816ed0e0b3bccc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"be2be34c0f3ddf3d338dbb71dc8a988ac3ee808f629ce8f45d7764889cc176c8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"5a4b3d4eac0e44abeabbb7303465cada2f191b4fd58367407378683fd95931a1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"98a0d9471d7ee36ac6f74d697881bb467bda26dcc2a07b5e4e5aa1cb48ffdbf8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"56c3f116152bd5dd351072172ec671849bd3ab26406b774cf7816ed0e0b3bccc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"be2be34c0f3ddf3d338dbb71dc8a988ac3ee808f629ce8f45d7764889cc176c8","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"73eea9e225b7b21b2665de8b7b69df84bad666b237757c756f5bee0a8ab15fad","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"e2566e851b7a4c5b17f5d438993befe0b0a52fd285edf2c839108a989e6785ce","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"a2c32d634ae898672a5ef4337bb5545e6bcf2060446c5be72eeaaf8fda78f56b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"51cc55562d3a56790ff8c8bde287c042a9500638003480a40e6343664054faa7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"691be2ec0e45ea7401d2d1fdf1685976a49e1d2d9b03b437a0c4a2a4ab4aa53c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"7c9ad8490a78b2b51aadc9b0b306cd550962084dc8a9ede2f42650a93444fc1d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"d9bc83cdfc00843fc74b975df41e780af87db30d90cbd377aedcd1f25f2ac2e2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"9a88b9cf27fffc25a6decc00addfdf923bd720abc0c3f95d94b5b46f81c86a7c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"0775d7b2e0358aa6e82bbda571d7a45e2d872df20c52f042bfcfd1219449c0c9","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"0abf0c64a9ead3fe4209902af9f489d83aaaa23c9e8e650bda1bc47250498fc2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"497c032895215ef8e7058610d20d0e132f267db9bc1661b7700f55c568dd8f23","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab8a90d71a2768c90b5158fa57f2395d4996031da73c96d28166b915f6c0f709","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"b1ea619032a41b4c9ef76c609451a8083e102c113b0acbb51dfde13e3e9e3448","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"f5626ace76e61bffc5312a61dd76c9f51f4d712a8196cf69b2665308c107745d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"c12577380db587b1b0b2aa784a1388684a8878fd82aa81586def46c91bfb5013","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"0ffae107db622234864447ed413809e48a65930884838b7e37340b1419874155","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"01a01346c2566b7c673847ad1cd41355f2f3cddefad072fe8e42ea5c848605db","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"c565c95da4865cd9f305e6ae81f6a8e0f7528b4e0763a20acc786bc5726cb459","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"8094bf5ad6e2ac4ce1f5f9945da90993c9a89f65564a4daa103700f686bf9dc4","bytes":5343}},"requestSha256":"846f520d697b1a44167c79f8852f3918ed52a45621d4a7148df51ce12abe494f","metricsSha256":"ea4c7f8eee52e1a32b70398ef5baeca7b829c2159c94bbdd6ed393d630630ee0","continuationSha256":"c983c876b3bdd93d1624867031d31c4856b57c941df73f197a2e5a22fcb0ee3d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"0abf0c64a9ead3fe4209902af9f489d83aaaa23c9e8e650bda1bc47250498fc2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"497c032895215ef8e7058610d20d0e132f267db9bc1661b7700f55c568dd8f23","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab8a90d71a2768c90b5158fa57f2395d4996031da73c96d28166b915f6c0f709","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"b1ea619032a41b4c9ef76c609451a8083e102c113b0acbb51dfde13e3e9e3448","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"f5626ace76e61bffc5312a61dd76c9f51f4d712a8196cf69b2665308c107745d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"c12577380db587b1b0b2aa784a1388684a8878fd82aa81586def46c91bfb5013","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"0ffae107db622234864447ed413809e48a65930884838b7e37340b1419874155","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"01a01346c2566b7c673847ad1cd41355f2f3cddefad072fe8e42ea5c848605db","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"c565c95da4865cd9f305e6ae81f6a8e0f7528b4e0763a20acc786bc5726cb459","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"e803a07058e973e74c0f4a253ab11a3a186cf0be3754754ced6f7590e8d784ef","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8dbec3e35359f2783ab9822db33ab167effb71c8414780394a77534afe98605b","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"31081c272aee34a76083bf9e7bd699f7687fa51b1bd59f696bf40303962ed24d","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2481821","recipe_identity":"4bbdc4414ee734a474c5d212a885011b38ba26d398b7ecc29b6288a363b62ab8","loss":3.509875353364249,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"3da063483d8b81406843a4efd5cea8ddb4181d906223eb406f8515fa1172890d","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"0a31908c73cae95186569779106aa69de80ec6aae4b2f274bbae9c5cb4e2270a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"f5d5bb139ad17b934ee28a020559b733662261dea4b6a3197c8111882b605cec","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"945e3036ba076521fd0a332e05a5ef5b84a4d464bdb903e6f2b28f2b7e259839","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"c19148cdb24552cff0b0c51b2dedd7e860b8f24e6e97996f50ac6e78be140af3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"cf912f35976f91a44bbaabecbee60c21771cef88fb52cd4835c5980e4ead8a74","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"31cc93329a2ed66430246c4d36fc38e6a2d0cf990173cca624b04a759c526b0d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"945e3036ba076521fd0a332e05a5ef5b84a4d464bdb903e6f2b28f2b7e259839","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"c19148cdb24552cff0b0c51b2dedd7e860b8f24e6e97996f50ac6e78be140af3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"cf912f35976f91a44bbaabecbee60c21771cef88fb52cd4835c5980e4ead8a74","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"31cc93329a2ed66430246c4d36fc38e6a2d0cf990173cca624b04a759c526b0d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"57aea5ff7257f308201eef4b066a4fd53b53f8119c79acbccc93517d1538d0fd","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"26be4bdd90bb27001ae955985e3edd8530ff500453a61126562874b7065b7333","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"9adc4048a1c8191770c13ca2d99cb83cda4db3d5e708a873c6668202ba631413","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"7623676d5b3d974e94bffe533b79e9a6e7d6548b40aaace015aa62ebd1a0fa62","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"9e58cfa58fd3e4890937fc325c51081680a9ada6ea5f1618010320084ecffa39","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"eb1bf44216f6243cc97ddea7be28ce5c871b4506daa4ba7632d5b1ec7318885a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"e53d827809494d008bd47370082b34a584d530947623555b42fb851de3902e3a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"c895311708f567e1823f1fab6a4b8b35dc5b4ee5bdad23f49d982b0316d3b657","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"ddd880dd77c6193e6de557c9a510453d30c937129ec15b5be95c39225d97a69c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"dda5c026e56b7eed1576ab682ccc3ba1c3bd66f4a0c4a3de049ecbb0ad4f5d01","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"3a2cebab9264ecaf8838e39d49f9599bfc7f44c9d6ac0b163bab6393de1a3358","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"f6b433b5289e50b736dbcd3356376fef4204254f8aa1ad323a3bd1f051feeb9d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"63a0654408e01c8c50ef5831bb916b56518f58db946bc79fc70f725040a5bd11","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"45029041186ea78dbcc9636b97884bea84c7a063f6a471225e1ad23a88253248","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"8db6a4bb491ffd94ff609946ffea156dc2accf747e3f2852dcafddd7a15c0bd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"cd65ad4ba2e6c59cebb2d21ca11f2dac5dbde956a4a5f3051336083812d8ffcd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"4006b74d4b0d1f522f9bbc08db63a5dda66a2c017815dccb080402f124ab2f69","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"906893cc76251d75f61308591e4b179ad22467a6c2d8755bcb3538b9736217b0","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"59e5a2d7da61dae757d30343f6a70aeb7c0e82a3309db4c9ac69a9498f1bcbb6","bytes":5253}},"requestSha256":"f34c819a2b6536bc8dc80456de1fda14edebbcd57fd75a1c2be4798be2a3655f","metricsSha256":"39de88ce357d7e1ee04660d30520f90b5db5cde03c3d8c6ccd99b07f51e0f029","continuationSha256":"e9d761d8fe915cbc2d173e9f5d9f637c6b396e61ddc7d09e6b1d36bfd905797a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"dda5c026e56b7eed1576ab682ccc3ba1c3bd66f4a0c4a3de049ecbb0ad4f5d01","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"3a2cebab9264ecaf8838e39d49f9599bfc7f44c9d6ac0b163bab6393de1a3358","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"f6b433b5289e50b736dbcd3356376fef4204254f8aa1ad323a3bd1f051feeb9d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"63a0654408e01c8c50ef5831bb916b56518f58db946bc79fc70f725040a5bd11","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"45029041186ea78dbcc9636b97884bea84c7a063f6a471225e1ad23a88253248","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"8db6a4bb491ffd94ff609946ffea156dc2accf747e3f2852dcafddd7a15c0bd0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"cd65ad4ba2e6c59cebb2d21ca11f2dac5dbde956a4a5f3051336083812d8ffcd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"4006b74d4b0d1f522f9bbc08db63a5dda66a2c017815dccb080402f124ab2f69","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"906893cc76251d75f61308591e4b179ad22467a6c2d8755bcb3538b9736217b0","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"da43ffcf28d886c54ca18f4b8e1a798fc1e2139637bd3c2d5cb35780e1d97f19","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6fe24bf5d539f7a65af01f604ecd2c1be81af9138f22af7d41f756c284425315","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4bbdc4414ee734a474c5d212a885011b38ba26d398b7ecc29b6288a363b62ab8","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s42","horizon":80,"attempt":1,"job_id":"2482106","recipe_identity":"c434bab36169de13bbad361f3fb762b0a7fc291068de0e1f00927046b77c1af0","loss":3.429070065939246,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/result.json","sha256":"13c9c3f8960b76c54642ddc189053b706ed681429840a5b316a4bdfad962afba","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.pt","sha256":"c2af3ccaf435b032f4dfec3698070c64edc87afc1d6decc63d0cc02035617361","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/final.safetensors","sha256":"6c3d85ee1d700ffa9c42655c2ef5b0d95ea90329e3c799e8069631e007d43911","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"9614a0271b28444a49c7ef10b1952558bee3a26b8076317894c88546a00d0c5b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"45a206983f434a459f578e676f701d976ac62ecf485cfc4815b708af2ae006cb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"cfab7fecff31337436b9383d118911d314ddc1a01769cce5af5e79b7beabba85","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"0e0be9230df98b3803832c49e9a097f597d138a0429a2999dc9b3c96460dbae4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"9614a0271b28444a49c7ef10b1952558bee3a26b8076317894c88546a00d0c5b","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"45a206983f434a459f578e676f701d976ac62ecf485cfc4815b708af2ae006cb","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"cfab7fecff31337436b9383d118911d314ddc1a01769cce5af5e79b7beabba85","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"0e0be9230df98b3803832c49e9a097f597d138a0429a2999dc9b3c96460dbae4","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-143.pt","sha256":"2b386ee53bbfce23146c6de5fef53cb3f78df43702eeda5568ef7aedc9c2be2c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","sha256":"cf8199b18a107ba585da705a6a9ba10a7126e63ab3efb9bfd152b2309ad17d3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","sha256":"e0c169753bf848b42162ba4285f24ee8a0ff605c8bba287f2f7bf948ea0e47a6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","sha256":"589ac30dbececbb46793973dc883a297b005ae1968b06cc3c83e7147c379f21b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","sha256":"a9e18246940389a07e1bb03a3843e372a8449903d2ccde5f4d11f9f699d75187","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","sha256":"6eef990a319e5378c069b0d35c757fc0109791cdff7477b6a578e1cb62dd99d3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","sha256":"0ab7694d5d3877e7efc03467f2cf9438a2dcf963e5740d7da177b18606d54a6c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","sha256":"c91f73a48d026039fcdfe02a61176e188c159faa72f6a202516b879fb31df5f7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","sha256":"93e3f5c586c7a10ce777033b4e3d631a401f4b03e9da38a1848b7f4fa121e418","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"dd11eeb6d02973b526f816cdc32c79ea7be447649a10332bbf839b00b0482f27","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"184de996f2a97a5d88b698e4bad5af65669d04769e121094dc9b03f5e9ea679e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"31841fcb8a7c768b086a06075f574cb7d277a99e5da6b72e642b0646d9656145","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"bf63aedd5eba627a86dc4fbf0c1fb6b4b2cadb87beaa9ce22c709b31e6c28d0e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"675fa4d9b70a7a31536209a8ec7883854dcc3f692b0cd0ed5b368e956bea1f62","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"ed863f2157148f5c110ff4ebd53916ed30bacc26cf9753ac18d82034c34d5ebc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"e96eae7e68f9ab7897e4b938eccac34554774dd7f32dcd9f5804b2827d8a8d65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"ffe98a0f9b0038f67576cc3eaad49802349014aeb0f109352770efa4b9c84887","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"28b15a2bb7d11780e4f9718c3ee998b30e7ec219724f497927e2fd3e7e3add6f","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"7e132f14e05e88ae4286cc93252095d73641cc767ea201e67728a7bb44cd0226","bytes":5255}},"requestSha256":"896b64878bc7e4cc7bcf153d0f7ce498173d6eff2487eb02bf84ab058273db11","metricsSha256":"ac657de0885b2d380ee4d32e40f8a851832733efcc2304560c2c847f17ec5d8f","continuationSha256":"9d69fe033ecddb500756adc543958888da329ef7c4c7cdfcb4aad38bf56888da","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"dd11eeb6d02973b526f816cdc32c79ea7be447649a10332bbf839b00b0482f27","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"184de996f2a97a5d88b698e4bad5af65669d04769e121094dc9b03f5e9ea679e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"31841fcb8a7c768b086a06075f574cb7d277a99e5da6b72e642b0646d9656145","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"bf63aedd5eba627a86dc4fbf0c1fb6b4b2cadb87beaa9ce22c709b31e6c28d0e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"675fa4d9b70a7a31536209a8ec7883854dcc3f692b0cd0ed5b368e956bea1f62","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"ed863f2157148f5c110ff4ebd53916ed30bacc26cf9753ac18d82034c34d5ebc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"e96eae7e68f9ab7897e4b938eccac34554774dd7f32dcd9f5804b2827d8a8d65","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"ffe98a0f9b0038f67576cc3eaad49802349014aeb0f109352770efa4b9c84887","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"28b15a2bb7d11780e4f9718c3ee998b30e7ec219724f497927e2fd3e7e3add6f","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001/continuation.pt","sha256":"9af9573bd6dd1907d3974246110313e00ae608876d626f40f135496ef056f4ec","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"021c54f018081dbbda4334d104054ea341460e292ad87a9e173395b6b3b382ba","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c434bab36169de13bbad361f3fb762b0a7fc291068de0e1f00927046b77c1af0","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s43","horizon":80,"attempt":1,"job_id":"2482107","recipe_identity":"197e54dddf64fee322942d02376a0d392ac7294001161ad0247e1fbc5e107692","loss":3.431562036894717,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/result.json","sha256":"511ec236dd71c02801d37c81af94cc3341421eed9742f1a78e0cb584879b9812","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.pt","sha256":"dbb848200a032320237a1554a5a8a7fea11d1094de09d8925a345269ed4381a6","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/final.safetensors","sha256":"18e0b59f9818fa00760cda05d0ae3a85250bb4ff78f0711c5cd2aa61b623654b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"d0d51a4d0dad9cc3fc965e181845463c27f78275c41533e6d74dfe7b39181ddc","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"3ce2a97795356165ec360f14f0dc3bb9bb0b49182942887365fb6d55197a50b8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"2b7395b00d36704e9e881ad14336fed906a344dfe104fbc667ebb1cc9f197882","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"4c7bf531ea53414c1368359d4a830e88f97588c165cfd353a41f8cfd39718550","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"d0d51a4d0dad9cc3fc965e181845463c27f78275c41533e6d74dfe7b39181ddc","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"3ce2a97795356165ec360f14f0dc3bb9bb0b49182942887365fb6d55197a50b8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"2b7395b00d36704e9e881ad14336fed906a344dfe104fbc667ebb1cc9f197882","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"4c7bf531ea53414c1368359d4a830e88f97588c165cfd353a41f8cfd39718550","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-143.pt","sha256":"f2b06f9777249b139a746376800b7f46bc419231d08b810b65487a2addfb5534","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","sha256":"9fd1b9ec93c79d6c5e9bc88e3b95299051ba544862c6d041acbfd0da80630655","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","sha256":"f795148d937e2462e033b07a2590cd281ff8f27f5392b9edb98b606abd16685f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","sha256":"3b1a53de5eaf0aeb61e793d9c140d6f7876f98685103d1499885ce65d8136980","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","sha256":"09613ef8e8c0fc34a254ee0fd6b8a5ca664a7cd917558c6a64e6259f5debfb36","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","sha256":"938355121bb9472aa37a8f28927df6e40b57b73d17091125e7e02beb4f6b9aa1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","sha256":"c2e1e97d791ad44c923cca1bb965fadf171ccf4d5658a0e708bf299a8284328e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","sha256":"e77d37df8af6478bd614076038605f64909bd487e05096d6e18e724ed84b3933","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","sha256":"af6488f06aa319cf34da724afcb96bfa95640bb1f0b2e6aaacb756c2ca883905","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"99ae3791b31dd8c3e4f9599f773f2243306ed7abbf12bf062cfe54ff88455e25","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"77088fa45d8fdc6bcece6ef662187e7ccfbd248dfd1bac8e09959d6f4ab62e9a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab7525e0afbd268e0f35556ed23feaa5de8ca582eca95f9af67d9237e7207d87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"71933303b01fb743b36bf4c7e39ff30e525c51167ec0d852a5ea7786a42f6f7b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"fb15417f8051810efdef4973448693e490983ea615f5ce6d89dfe20dac2e622e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"cf67881a46d017385bd1ae3763ffa76875e94cb6f8472d83e6c28e3b800be2e6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"dfcbe83e6ccb2258b61afe6912b6933cebd19e71ec61608cb1408ade80941606","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"f5a84d9e6144b9b39e6a7541a00e9c77c43603523709bc48c9035b31d3e9669b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"5d03d9caa8adc0d68f252ca325e8acbb4b3fc27fea4f3455c89f848d35293378","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"4ec245095308140517c1498597c6629313f1c4f88ba5920f482c2c35d0b88d9a","bytes":5258}},"requestSha256":"6ba7f2e830773b535cd347df568970132bd0f1d40dd59c58267358ca98356789","metricsSha256":"6b7b1e8bc9a16d53c663aadcacc03f6b9e5c9a8a5c991a297f71a0b37eff1e42","continuationSha256":"2a78fe9d1037d522ad73cd4c8a468d21dd6ddc6dc210366c0aeeee2ff804cda3","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"99ae3791b31dd8c3e4f9599f773f2243306ed7abbf12bf062cfe54ff88455e25","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"77088fa45d8fdc6bcece6ef662187e7ccfbd248dfd1bac8e09959d6f4ab62e9a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"ab7525e0afbd268e0f35556ed23feaa5de8ca582eca95f9af67d9237e7207d87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"71933303b01fb743b36bf4c7e39ff30e525c51167ec0d852a5ea7786a42f6f7b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"fb15417f8051810efdef4973448693e490983ea615f5ce6d89dfe20dac2e622e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"cf67881a46d017385bd1ae3763ffa76875e94cb6f8472d83e6c28e3b800be2e6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"dfcbe83e6ccb2258b61afe6912b6933cebd19e71ec61608cb1408ade80941606","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"f5a84d9e6144b9b39e6a7541a00e9c77c43603523709bc48c9035b31d3e9669b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"5d03d9caa8adc0d68f252ca325e8acbb4b3fc27fea4f3455c89f848d35293378","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001/continuation.pt","sha256":"85ad5dd77d0260fc137367b0d53c0e67af604725514b23d512d85305ad715c80","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2f282ec91df84c979b87c0a2dbfadf9fcb4e72c325287d21a9685a34361d9fc6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"197e54dddf64fee322942d02376a0d392ac7294001161ad0247e1fbc5e107692","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.006-s44","horizon":80,"attempt":1,"job_id":"2482108","recipe_identity":"cb28c3e43fd58a885eea8c6f01f60f25115f110938379d8ed68b674122dff8b1","loss":3.4307619388913335,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/result.json","sha256":"cc186fe721d63c8267539d1f3bfa0f571ae46d679744737abb60676a992f2aae","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.pt","sha256":"dfb9278addde38a9c4bdcc44b5ecf992e3cf18ee30e18651033eacb7f4a95828","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/final.safetensors","sha256":"1b8cde7af7b2f74b5f6e2f6bf21dd5f47b51d7817c8959418dd04479d06c0e41","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"ddd3252b052f4e190dba1a734c57e9feae0960d8eab706bbd2cbeb5ac16ce1c8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"974f870924272e083de7489c17238e958c275421df0bd98ab5de82cfebc71fff","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"52cb793d44c0bc6e41f3815d761cc70e6b92536f399faa337742b5bb31f787a5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"198e1d7dae54152dd8aa141e4c5c1dee5035141f3397ea93407a40771e5cd4b7","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"ddd3252b052f4e190dba1a734c57e9feae0960d8eab706bbd2cbeb5ac16ce1c8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"974f870924272e083de7489c17238e958c275421df0bd98ab5de82cfebc71fff","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"52cb793d44c0bc6e41f3815d761cc70e6b92536f399faa337742b5bb31f787a5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"198e1d7dae54152dd8aa141e4c5c1dee5035141f3397ea93407a40771e5cd4b7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-143.pt","sha256":"90c499c0b39c951ffd20bd700c1f53db0d6192ec971ce1f928e9e1599097da0a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","sha256":"5011f29593917ddffdb685191acb0e7db7176fc3ba604d663cfcd2f4cfd3730a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","sha256":"9a35e68b606955cd011aff33d1d0d828b44d6d764813050c682fc0a7d1f699fd","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","sha256":"e0df70cc23f22d294b2f0adbbd7e3f63ac8c1b77ed9ac067f95b15dc0a43e947","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","sha256":"410af145fb2a98295a308af37850371f6bb4ef5b546f03984c9a39d3a8c56ac9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","sha256":"cf4fc7c5cf9c31aa6c53a5d9355b896e878d3771c16a4b01528005208e64d4fc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","sha256":"348b3107ae68bf0c796ba2318d9aa376667c635e295fcb64145d05e0697ebc2e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","sha256":"4e15f231895b80a3f6aa6a0f91dc3df59e89a877047cbfb16227aba96c015830","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","sha256":"41d591018922938b57ed24897b27b99051d849d153133b8aa9c85314f539d8a1","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"d1294ae44a9861abfc4221a86f258693fb34cce5184f6a46d768f63e2faf1c20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"27ff53907cb7c138a7ecd7abc972f85b9f7d41f189c3378fbd543e6975e7bc09","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"bdcf81e33b2331e5b935413167c5c0bb747867e95d1542a10ba10443dae95ffc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"af89e8fc4e5446e84bc95a943e7446e6dbd368f22ac349fe20ff52fd2c63ba3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"c6247543e999b64af0414a7d02e6f164990e3e696e6703bd23b30ef37fda457a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"41efdc84a50c26a50576cfd045b4e24fb3f36eaf269795823af2cfbe110dac8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"58151732d953d5088d43e31e579f8709788913023374800a451124d34a91fa0c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"b952477afcc011286a04b00243f94afd54a3ad83156bb19ad4b4572357e51bc7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"4804360e7db1cab2979b2af59e8792279b30d5b29ee8e0adaf14219cdc3b69e9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"b01c074a85c40c33af047cd336e344c65616282148788e23832bfd4ea26e3d23","bytes":5257}},"requestSha256":"afbb017fd23d6d2749836f02a1ec71228a9eab3814ec98361ea6f8dda9d2ed5b","metricsSha256":"080bfda688ca8f21926d11ae00b6433496244ff59f5c2a29f45cf8bb4467b2c2","continuationSha256":"fa088f4950541b026b2796c2aa20168d3050fae54dca934331a8d154b4e53dd9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"d1294ae44a9861abfc4221a86f258693fb34cce5184f6a46d768f63e2faf1c20","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"27ff53907cb7c138a7ecd7abc972f85b9f7d41f189c3378fbd543e6975e7bc09","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"bdcf81e33b2331e5b935413167c5c0bb747867e95d1542a10ba10443dae95ffc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"af89e8fc4e5446e84bc95a943e7446e6dbd368f22ac349fe20ff52fd2c63ba3e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"c6247543e999b64af0414a7d02e6f164990e3e696e6703bd23b30ef37fda457a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"41efdc84a50c26a50576cfd045b4e24fb3f36eaf269795823af2cfbe110dac8a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"58151732d953d5088d43e31e579f8709788913023374800a451124d34a91fa0c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"b952477afcc011286a04b00243f94afd54a3ad83156bb19ad4b4572357e51bc7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"4804360e7db1cab2979b2af59e8792279b30d5b29ee8e0adaf14219cdc3b69e9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001/continuation.pt","sha256":"495c495e7f7939ae56d7ab6b153cd07a9352dba2d22e5663ff7adf787fa8a761","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3f71596b1cc4d6996d7a4ee25d608392b7f74b62127556f45fd94ccbfeb453aa","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"cb28c3e43fd58a885eea8c6f01f60f25115f110938379d8ed68b674122dff8b1","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2482109","recipe_identity":"73ff79d52cb6fcdf88db094defcbe3f092ba3d9aea2af07437b1fbd35a6f5e22","loss":3.4282375474222753,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/result.json","sha256":"42ed2244101516563ab5a6663f96b49ca975a67a124ce68886ff8441f18a6286","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.pt","sha256":"3d7d6cb1cca28ce82023d7bbb644c7a37aaf6749194ba172b915fa5da02db939","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/final.safetensors","sha256":"159e04da2d266e4a9ef08ee99f727f36a6201f81d0877be9619a60c711c61829","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"f7a86ed2807928c2ec7a3baf15c8a951f717366eb8f9d5c867c0e33026cd0fb0","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"360efc065700f23354d504cee6e1bac4d6add1abe305f9a483bb6264a0733836","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"c31ff1bb55491fb1eeec75693c9859c37c068d752813a128cf2583c39c0a9e8a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"a69cf8376bfd9c87f81a42eb3d98c1939f03068834be2276180cb7e7851fd048","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"f7a86ed2807928c2ec7a3baf15c8a951f717366eb8f9d5c867c0e33026cd0fb0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"360efc065700f23354d504cee6e1bac4d6add1abe305f9a483bb6264a0733836","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"c31ff1bb55491fb1eeec75693c9859c37c068d752813a128cf2583c39c0a9e8a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"a69cf8376bfd9c87f81a42eb3d98c1939f03068834be2276180cb7e7851fd048","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"bca81b1990a1b69bc24603573c7ce19b4a446d8535bfa739f8821a7f1d22ec3a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","sha256":"6dcb01ca57b39c15c2804920483dccf81bba3cd8f419377b3b142189f06ccbb3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","sha256":"6d036f801f4406f666754344c31b1b993e99f25a1c0054347e15985f531f6d35","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","sha256":"6968928387b092c72035d7cc6df62bcf4c5a082dab10805ea2308c219bfcee15","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","sha256":"a9cb55d434688133eb0236ee810e2f0ae4cbdd788a454a34218cc63411015007","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","sha256":"8a771d58acabc9a65c5b33cd5212d81f6978e5a9e17f1b855ca1f1b8dca4bd53","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","sha256":"402bf09ccd85b737fa34ea4cd98b8d789919b161d71029c2561b96b5548fec59","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","sha256":"06ed1d13af02f8bfbddf357f4a71f275f94e915f7f0197fdd455b8225508b1ab","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","sha256":"0662379b91236cbdf45556faf5c7c18a9b7b5382629b28150558f9f1d735db31","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"3d13025a0a3908ee3b3f758bcb1453ccb8a8112e025d55c26a75aa901764f3a7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"50cf8160f4765c8250d04eeaea53d9da61298fbdfc5720502328060ae65b3822","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"83241f5276cddf8a145d5994cfc352c4f5fc4749461a657971f281c15a944dd4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"faa4c52f7dd49c5641a1383a6e9e12e86b9555fa2a71898787703d808ef583ac","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"06e48d160bab0eb2f76dac5976a108ff2b66163eb56f7a455fac43dda1aefa3a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"8629f7fccad833a64c73be90c05116cd9d43d18abcffbc2bdc786c4d3786725d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"33d2ce68160563bfc3a0df22d1f4e073a70a0feb960ca89d9d61377a5bdf3b20","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"67fbc4178cfa00e5b2033afaf2f718f5253b4e79a17674ea11b9c31ac369af7e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"061cf029e03bcc4e36aa5e48daf1b570352b99f4f0c54c3ed80e9d965eb5e9a1","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"0895734e7096b9ef2d3767ea809c76dd6b3cdd8b14bbf4564b2d963b7e11cfb8","bytes":5256}},"requestSha256":"220e6f96322ea19559e3538355bff2ae8c7124cb2d9472e5b9ea65a1d6b1a747","metricsSha256":"8d131d9c81fc3ffa8366d08cc42e698fff170c94ae096c48c361320c457b4524","continuationSha256":"43c4695db5d37647816657b64b56340d1a927e1e5cadd7277bab88f61afeb097","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"3d13025a0a3908ee3b3f758bcb1453ccb8a8112e025d55c26a75aa901764f3a7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"50cf8160f4765c8250d04eeaea53d9da61298fbdfc5720502328060ae65b3822","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"83241f5276cddf8a145d5994cfc352c4f5fc4749461a657971f281c15a944dd4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"faa4c52f7dd49c5641a1383a6e9e12e86b9555fa2a71898787703d808ef583ac","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"06e48d160bab0eb2f76dac5976a108ff2b66163eb56f7a455fac43dda1aefa3a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"8629f7fccad833a64c73be90c05116cd9d43d18abcffbc2bdc786c4d3786725d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"33d2ce68160563bfc3a0df22d1f4e073a70a0feb960ca89d9d61377a5bdf3b20","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"67fbc4178cfa00e5b2033afaf2f718f5253b4e79a17674ea11b9c31ac369af7e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"061cf029e03bcc4e36aa5e48daf1b570352b99f4f0c54c3ed80e9d965eb5e9a1","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001/continuation.pt","sha256":"2a70e4860212c4ce1d6f01f8d83c8b3f33e5c36e0b4e9e3bcb313cfbc5112122","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"909400517bf7ddc62282f2ad1d63e5fbf0cc91b2ceaecbee3bd13e1197259bb8","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"73ff79d52cb6fcdf88db094defcbe3f092ba3d9aea2af07437b1fbd35a6f5e22","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2482110","recipe_identity":"84b524d14bacc8e89ca530a64b7658c90be8c5a41d55763d67695b4508f5830c","loss":3.4317096205957034,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/result.json","sha256":"e646a6b4db268f3de1b0e965f7f82f8f2262d83c4cd602dd7604b1c3a6f30325","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.pt","sha256":"eadcd4db95ab046ce78212dba9cdc640fb01141f8d317bb296115c468aa2c354","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/final.safetensors","sha256":"fddcf952bc98c7c45c2316353019f2d67e4627bdabc5c35eb5330e6b91d5b35b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"4a1d517381ce1e615722f8cb93fff35f2f2a33d91526c398b969fbf242a09645","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"5653b1ea5f0ed442515feedc95e261bed3a9e0793fe3acd822978a23f14f71b9","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"1429ebc8c7a902180b7ffa98e11db2a8aee9d07f551b361fc5a9956403ac58ad","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"f0e319aeaaa606fce2b33de9faaa349f71878e96c7eb6986bcf7f1882c0f0613","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"4a1d517381ce1e615722f8cb93fff35f2f2a33d91526c398b969fbf242a09645","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"5653b1ea5f0ed442515feedc95e261bed3a9e0793fe3acd822978a23f14f71b9","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"1429ebc8c7a902180b7ffa98e11db2a8aee9d07f551b361fc5a9956403ac58ad","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"f0e319aeaaa606fce2b33de9faaa349f71878e96c7eb6986bcf7f1882c0f0613","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"33bba145a8a0640f7d0debd382eb5d927ad05d76639806c1d71eb62ae0c0a652","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","sha256":"d05787210a3b71f1d09a00cc04e3d7d84c5f14cff76e7cf4ccce5f47d27951ce","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","sha256":"1798eb85e4ee31ae02826e4d8b6b36f757f0a8808d74977111517059b61e6ce7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","sha256":"8c93d03b015a58db908da941b9f13693a760bdc5e7280addb621192e477cd0cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","sha256":"4f398017db063654224feed1fd4c143d09adb8cf6a38fe400ae0f7e9ef7490f1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","sha256":"ca6ade1975b613650fdfd9e5003f86a6deaeff37c48f2ee6632e7d7e36063d7b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","sha256":"e144d77c9d21aef5381fa631712b862d198b93dda3f71c50e9dbc796b64eb5ee","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","sha256":"ad091b01c3c66886ce3f16eac3641a6754dec3edf0752bfaa77e9dfc8e5d87c7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","sha256":"b161f8638eef4fe4f3bb2e5b4ea35dab52756e6beeba2f7aa8757f59457e948f","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"8aa0996a2631f75c78233e141f1f81148a67e52e6b873ea72ada8decefd50660","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"5a7a0dc44ecb4cfa1592536d711be626ba927916abf084f975b0474c9f6fc8e7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"321532ffd17fc8e84c5e4c1d33447f24cf4f76d31ed93881b32c848f20ac2336","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"d65d0103d8731a894893f6b349e278a56fc621ccd2754add4dec3b532961e41c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"e7ce173864f848932cf5586022e45f680bf829ebc7a5a62a883c369a46025032","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"2ac129af3232491fe1c1a996ffd8ad23d7369854195991f2912815a5fab7deec","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"e2ae409fac374cb93175277f72533d5dce58fc432a362d68950ac0d1bf402d9e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"729766b42979aaee2958c52f3cab0fa741341ad95303585cd4893f93c8d14f58","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"cf43d06303a2e12652405d2c3d4b408299ebec57fdebee628d11f720551ccb52","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"281e1ed1eef1fecfe3c19ce06157e152663054396bf04704f55af495d5ee8c0d","bytes":5342}},"requestSha256":"06bcc6a1bd7f9894d70f4cb06d7aeaa90800c4a84c1417e54081f30e6d9cfa7e","metricsSha256":"7408f82830518935c7f40900654600596273d8149bab9e896bccb8f838cf0461","continuationSha256":"18f3f3446698a098bc9a5bcc17b7cc4ecb41c52c47a605a02965cd1aff7d031f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"8aa0996a2631f75c78233e141f1f81148a67e52e6b873ea72ada8decefd50660","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"5a7a0dc44ecb4cfa1592536d711be626ba927916abf084f975b0474c9f6fc8e7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"321532ffd17fc8e84c5e4c1d33447f24cf4f76d31ed93881b32c848f20ac2336","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"d65d0103d8731a894893f6b349e278a56fc621ccd2754add4dec3b532961e41c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"e7ce173864f848932cf5586022e45f680bf829ebc7a5a62a883c369a46025032","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"2ac129af3232491fe1c1a996ffd8ad23d7369854195991f2912815a5fab7deec","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"e2ae409fac374cb93175277f72533d5dce58fc432a362d68950ac0d1bf402d9e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"729766b42979aaee2958c52f3cab0fa741341ad95303585cd4893f93c8d14f58","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"cf43d06303a2e12652405d2c3d4b408299ebec57fdebee628d11f720551ccb52","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001/continuation.pt","sha256":"b8bb356786f3d9708509d89ae087315fdad07240051ef008a3bfacab125f74b9","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"082147df37807ff31bca126f8782c53f2024960b7a167f460f75ef9fb323a5fd","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"84b524d14bacc8e89ca530a64b7658c90be8c5a41d55763d67695b4508f5830c","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2482111","recipe_identity":"6c56d93122ce190e094bdcc52e7d33b8c324e6af0f0c6de5f2f75275340d2235","loss":3.431566621085528,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/result.json","sha256":"194b4b230358b52d377547a06b262fadf90d6ce3b5ad89def363efdd5267539d","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.pt","sha256":"c7c22eb08b2690d73e5f55159b6734463fab474186b91b2e9f7f5786c677ea5b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/final.safetensors","sha256":"3aa83aca07fbf0489af9a408ec5f008695fc26fb15fbabdb200009c59f7ff317","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"1a3100135d97dd2092e49fbc6eff7dd2c0fb2b71ec7f85a28db0e75deeff2149","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"b6ffced9f31d216e7a3e4ac01bd6f1a5d0c00d9f70324822a526ff500b2155e8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"2830f665ec47989cf9944e2fe7ef0131ba5033cdee30393c3d334c2701c477e8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"869524c831492cedacac9a7052db7aca2823f19e0d82c5fcc39e57c6024e31db","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"1a3100135d97dd2092e49fbc6eff7dd2c0fb2b71ec7f85a28db0e75deeff2149","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"b6ffced9f31d216e7a3e4ac01bd6f1a5d0c00d9f70324822a526ff500b2155e8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"2830f665ec47989cf9944e2fe7ef0131ba5033cdee30393c3d334c2701c477e8","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"869524c831492cedacac9a7052db7aca2823f19e0d82c5fcc39e57c6024e31db","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"5ea1800d07ac82b396000b939fae852b1919150c3d5071ebb817ba1eb13f9dea","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","sha256":"22dfd4a98692a33727b9dd2a70c3e371ca0d206ceb53988451d9e4207becf777","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","sha256":"5b94e16c0d49c8c0a5102c2bfcb3121cf2f783387224e84eaf06aff921320558","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","sha256":"b072f5221b8989ef43a950f810c1d2ba2bd552d6fb76a8dc09cbb16201a3c497","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","sha256":"b3a76843e9c42d089cdbbcce1feb3f12bd03873d8b5a12fe5ae6df0e895b8ae6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","sha256":"2843b42806347e6a9a8b1c42fd6bb7aeb175dcdba15b2bb10b2a581d00d4fd24","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","sha256":"aa190e48441bf1ade4749c6d3bafe0604f9ff48cf32d6baf3bc7bb23d4e0832e","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","sha256":"940bcf797e5d034266ddd54af2b878dc98954ef1b89075ceb06dfab1ad53898e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","sha256":"eff27af843880eb4c0e6e79a7ebaafdb0ef874875bac550a94d7e54a857157e8","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"35e19b6a0bb6188724831bdfacbe6b0b9b9df21247240e28ad0368884c1dc38f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"eff43d733005d8d70c2ca3ce55b260f5671ab5226b9db481d7449cbd69355b7a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"593e995f6fc3822ea54d86b6122dde81bfb3b11cc950f750804b82509dc2eff7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"4e4868caecdaffc71a216a64c8c1242fbd140bd61d50142a33b78996485187e0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"e3174fda7989b5f9bb06e416812390a26902d3480b778a232e17fe433f23a625","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"734e9687a5f38ac62de7d01821a5fcc26039d698364d2180e4f18d4dddbe3f5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"ef497ee110a07aec0c401f86243c86da46561191a55ab7cd381beca9754cde00","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"dfd41b27f9f66be780b426d4b8810a04184d4eaa74cf4e5c24c7202bb1790448","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"be6aa8d2426fa4c7f4896b429a7a27137d4f45e10efc0fb30faeed7b015bf2c7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"590dec7a3c69c56fc2bb85ca0373e2bffec5fa5c646aaea9b0354a8549c6f5cc","bytes":5315}},"requestSha256":"e990836384296e183cd015937d0a39187e440f40c25903fee3b6606cf1b8e7d8","metricsSha256":"1074769a7cc8c6e2edd574e102704439caebe7b291db7743e79a373a8153d664","continuationSha256":"ebed71ed80c58caa551782deb08e25806b590933cf79551daca3dc072620d612","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"35e19b6a0bb6188724831bdfacbe6b0b9b9df21247240e28ad0368884c1dc38f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"eff43d733005d8d70c2ca3ce55b260f5671ab5226b9db481d7449cbd69355b7a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"593e995f6fc3822ea54d86b6122dde81bfb3b11cc950f750804b82509dc2eff7","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"4e4868caecdaffc71a216a64c8c1242fbd140bd61d50142a33b78996485187e0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"e3174fda7989b5f9bb06e416812390a26902d3480b778a232e17fe433f23a625","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"734e9687a5f38ac62de7d01821a5fcc26039d698364d2180e4f18d4dddbe3f5e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"ef497ee110a07aec0c401f86243c86da46561191a55ab7cd381beca9754cde00","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"dfd41b27f9f66be780b426d4b8810a04184d4eaa74cf4e5c24c7202bb1790448","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"be6aa8d2426fa4c7f4896b429a7a27137d4f45e10efc0fb30faeed7b015bf2c7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001/continuation.pt","sha256":"7031d546730d3afb364ee6989bb4d1c7bc450ac031450d1d9f209e71574ca02f","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"92c453d654c0fdbd4946631bfbfc854fa8d0d2fe44309184590ec7d5477fd1e1","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6c56d93122ce190e094bdcc52e7d33b8c324e6af0f0c6de5f2f75275340d2235","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2482112","recipe_identity":"9685eb3d5e8d5c50551f5eeee2731bd7263f0af96e8469299c84c49ac7a7f32a","loss":3.425987926326014,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"53b6b7ec4439362c069ce7740c357c984fb20e7477e929b9b36143aaf06a7513","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"eecfe836d70b5e7dd132915162d3d6e102c3cdd87f47ae90d6ae1d67bd8941b8","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"d06a05b2c7916a90bfbf9d8ece165e505fa031969ce861a247813b29335b4b07","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"c2d68afbec83b0a8421206932f789963b7ee05411f9be39e6d6f9caec36c4107","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"a7e1e53bb96ac3bff98b36c0e1756465ddcd713723af9bfcf95c8d4d97f98da8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"00b8956781f8b40c1417c035c5c45efb079a7d4253ffe2020129f7b3c4ba66a7","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"57594ada5422934a9342be4f1be2ba7099df6071d35f699ad288738a6de2260a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"c2d68afbec83b0a8421206932f789963b7ee05411f9be39e6d6f9caec36c4107","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"a7e1e53bb96ac3bff98b36c0e1756465ddcd713723af9bfcf95c8d4d97f98da8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"00b8956781f8b40c1417c035c5c45efb079a7d4253ffe2020129f7b3c4ba66a7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"57594ada5422934a9342be4f1be2ba7099df6071d35f699ad288738a6de2260a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"1d23e2c2b60c8431f004519238405b7da776ab0c4db46739c7b9e0184b537399","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"23c0372b995794b48368e4325cdc5fa20fa92f6b02d54bda0533b4e3ed03cb8b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"c0f0082f924a4d761e4c727d930bac2560eced4a4a42d8255925e558d7145161","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"45b2d8c1e2e08c28b23dfc5113ebdfcb20238aee76f9ac58dd1f8f63015795ee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"f414c7f719e087fd20274a47e5d84ce7c63d8ae343ae142f5a10a56d7de38e5b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"3bcabf3c6163d0bb08b3d7434449b2535440735d7c4a8f864d258cc643a4b930","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"e11167b0c041026eb1f2fb8f0e6d6c6f9586258543b46c9276c186d1feac1736","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"c27a5f3b50e023154033295cf4c3e84d9bbe4bd1d833e95fa5828728093bc0ad","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"06bd206e4ae2c63f319b224b1193b5b82aaf73c2420fa017dafcbcde33bc7894","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"21a6b4d1a7259878afca38a141acf9572716ed6ab7d59d5a4a24dd2c4409a7b2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"4e4c8506e9eff24e1ce1eafb874ddaebccf0b1eb4cfafac30a08c8ad1ba21f25","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"77bb1ab5769728ab0d00b9a7490c1214f876b76ab374d47e57263177f0da2751","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"f29a3f1201aad817a6a93e63fe06c1998a9460f50cf22cf56ebdb20315b7fbeb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"a10b4983eea35b25371442961fc898caf3208025a175adde61faed467fcbe77c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a719ce6e6fb1ab1aa5ef60cab7ecd0fc3194cd1ad042b5bd9ece2959a5823c2c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"648faba86ddeb87f2fc50e6d82a59e5b7a8e659dfbef5b91a462913f7c67af84","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"dbc275df46d851bb41aa6d42434767baefccd2da2d0cb472f94423d699bbdc02","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"e72c268461bc09dc8b643e2a1edc91307a4e5cedbe9e53ed944a2164fdd6dc83","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"6a75ad3a0045a6e3127ec6ed7a666ddf11c76795150ac9e30468b4ca5dfaf426","bytes":5258}},"requestSha256":"54cefe1b69d28e73bfdcb3fd21174e075475e256234c15ad4dc774c9c754f154","metricsSha256":"84e1eae114ddbae4c72ee6a4ceb0a0a699ad46fef735814fbfe5a5a411d7814c","continuationSha256":"17ba75a00a91b79c8b523d1e98b9a85f68dff8c14b50a0384ad94461cdd88365","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"21a6b4d1a7259878afca38a141acf9572716ed6ab7d59d5a4a24dd2c4409a7b2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"4e4c8506e9eff24e1ce1eafb874ddaebccf0b1eb4cfafac30a08c8ad1ba21f25","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"77bb1ab5769728ab0d00b9a7490c1214f876b76ab374d47e57263177f0da2751","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"f29a3f1201aad817a6a93e63fe06c1998a9460f50cf22cf56ebdb20315b7fbeb","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"a10b4983eea35b25371442961fc898caf3208025a175adde61faed467fcbe77c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"a719ce6e6fb1ab1aa5ef60cab7ecd0fc3194cd1ad042b5bd9ece2959a5823c2c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"648faba86ddeb87f2fc50e6d82a59e5b7a8e659dfbef5b91a462913f7c67af84","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"dbc275df46d851bb41aa6d42434767baefccd2da2d0cb472f94423d699bbdc02","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"e72c268461bc09dc8b643e2a1edc91307a4e5cedbe9e53ed944a2164fdd6dc83","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"06d4ff5f41104d1bbd722c734aec9fc214d861ab2fc516a4935e060319f1f239","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1742112f5e28acd09d94ba266a91758d5adfa0b4ed52209df512d82292f23ec7","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9685eb3d5e8d5c50551f5eeee2731bd7263f0af96e8469299c84c49ac7a7f32a","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2482113","recipe_identity":"208df82acf96e75064e44629ea708657712f17990880b3896646b07865841da5","loss":3.4343659598054344,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"fceba83751e339a5b48d9192d77418bb1aeb607c6d0cb7b0a5656b11ffb83a25","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"51a7f418196e71ed55461683fe0ba9f774f5c639ced6b82599d3d8ef6c596968","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"2d87c27c17056b7334e582e9420b1fe04b4f9607753429cd417acdf2435035f6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"9ec00daa4759ff58045db7e4860e72e54513997c84206241f56a5d7e3d223279","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"6d4a223b05eed007482997332613a52f78ec4954add3f4233fc40aac0cae4a0d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"f882e91ca715faf8c5170a43924482010f5a7ee40eca6d4a8b447174c1babc01","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"801fc3287dd8c669d5f4bae89fd8a6de4e222ed172ee7b239b0944a581ca8f72","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"9ec00daa4759ff58045db7e4860e72e54513997c84206241f56a5d7e3d223279","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"6d4a223b05eed007482997332613a52f78ec4954add3f4233fc40aac0cae4a0d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"f882e91ca715faf8c5170a43924482010f5a7ee40eca6d4a8b447174c1babc01","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"801fc3287dd8c669d5f4bae89fd8a6de4e222ed172ee7b239b0944a581ca8f72","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1eaf51e19ba7fc6b4348c45c9910e18dc04e22ca2595e1de4cbb9335c234cbe0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"57231d2193a67fad7d2b17112636994698a88b931b6a9d06faee930e4fe9b399","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"1ad2c95b9fa684d620f237840051d9552cf0b3a99f82eb7e470045c155220287","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"bcbaa9c12706e92f0940920fde03b42ff3b3cbccacbc1a2618420a8883cf8edf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"3f765e991030d0b739a411566006663ae5ab7742ef28310eb51aee2fffbf6f07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"e56174fd8693ed7744159bc410f37841a66424acfd3de09aadec256cadaa905b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"3a7dd06abf27a6ae32a5938c99865d79f8553637cc17a88fb72109c5310589c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"798f5febbb37fc11ac967c6766d30da545075b55dce3341cfcdf62c21d4cd310","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"419af95fee81e750144f357e37a962f375648d382b565dafeccae3b41ee71fd1","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"aa7b6ba0743d89de2fee244348f67740028ff89d139a33b52b2a2f9eecf79ec4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"ad2305cdbad9f1e38157f618d770f1feb8f69b06c48509f8ec2c89c3d89ebedb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"de56db39ab066f31d02f03051e4d2e886e81ecea65b2f3282a8fad1c0599100d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"5d0871a2199ef53a29d63b16a3536633ad0880405564c2735779bf433fa391ba","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"b4f0cc5fa958ce504d8f04e05dc270301e934f546ed8c8c009a1b518b4185a83","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"623a8ea4f5852a4dfd4b5be868be69c4e5019871ac727f8728d2e028e63bfba7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"fbbdb9dd0a3329477adf020322454b4836d118f21645039f68b686a3f9f762f2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"5c7ab15639116cb50f8b76258128f9e72d819e701c7f7f04c6a38a1620f6616b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"1a008a6deeff4938ec1fe46c6a7c9210cf17a10234fb3263d5e3a8425bcfc74d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"0d6d6e50dbd76da244ce1ddf4ff70aa6fe8df876e13c10f8cb9adb55ec7f2a16","bytes":5344}},"requestSha256":"d22303f866bd93a9e3be194d5541fed3dc8c62ff0d7bc1678b64820a6ff11550","metricsSha256":"a158ae48504c2cbbc67abede987a323b558ae347d690e913e36c7b9645e6c700","continuationSha256":"0141063716b30dcd9dbaf411a5e7574ec4ad086ca687612eda6819e3258f7c4d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"aa7b6ba0743d89de2fee244348f67740028ff89d139a33b52b2a2f9eecf79ec4","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"ad2305cdbad9f1e38157f618d770f1feb8f69b06c48509f8ec2c89c3d89ebedb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"de56db39ab066f31d02f03051e4d2e886e81ecea65b2f3282a8fad1c0599100d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"5d0871a2199ef53a29d63b16a3536633ad0880405564c2735779bf433fa391ba","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"b4f0cc5fa958ce504d8f04e05dc270301e934f546ed8c8c009a1b518b4185a83","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"623a8ea4f5852a4dfd4b5be868be69c4e5019871ac727f8728d2e028e63bfba7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"fbbdb9dd0a3329477adf020322454b4836d118f21645039f68b686a3f9f762f2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"5c7ab15639116cb50f8b76258128f9e72d819e701c7f7f04c6a38a1620f6616b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"1a008a6deeff4938ec1fe46c6a7c9210cf17a10234fb3263d5e3a8425bcfc74d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"b0041ddd5bad595a4fe8e0d1c2631aa7a5a752787ec317a8cd1ab9dc4df81aa5","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"eb8853a22264bd174573eee50bf8d7aa9d960fc095673061c41ab126f8f2a331","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"208df82acf96e75064e44629ea708657712f17990880b3896646b07865841da5","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2482114","recipe_identity":"66124a8942fa1e4b467eb3132973809dbefde0a4f93e6399462369a9a8c3dcc8","loss":3.43147567038545,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"730b732c2de5ed5f9080f7689eb8d34756967029c88048fe9116914582efd224","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"a31edc811c1380f535dc368c694f01da3493984f066cb6c5e302604d8d2f72ad","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"81686057f938c33a9c0461674554884d51fab78a5f1f7a1869c739ba9e10444b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"fc94006d583c9bab226d99a6ffbaecff4a41874dbf01571664a1b4824ceaba11","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"172380feb23f77f1c17167c61cf75c854bdbaddab3f6a97fb23180a8b1dcce4e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"f1271916096f53dee3a077315f94caf352b41ddf04fb39d02d12a861f648454c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"e8f852d2a2d13faf09165c0cd26fe6dca30325382e5792d9253dc6b9598ec2ba","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"fc94006d583c9bab226d99a6ffbaecff4a41874dbf01571664a1b4824ceaba11","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"172380feb23f77f1c17167c61cf75c854bdbaddab3f6a97fb23180a8b1dcce4e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"f1271916096f53dee3a077315f94caf352b41ddf04fb39d02d12a861f648454c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"e8f852d2a2d13faf09165c0cd26fe6dca30325382e5792d9253dc6b9598ec2ba","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"de2d73618f9eea0e057e3944079e217f4a12642dc0b66853749efbd57cb6173c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"651f2b711b5115bba3cede325fa34d1b4d3086e6a5132f645c9fa90385a206e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"22710dcc178a9d673fc0b769e04e191b290189a80532943daecb792a3e3f059e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"043b0c2175117b84db68b0ccf1a625eed34ce4b769a314713cc11355ddb7f891","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"a09b8d984213ab4209c8ce650ee1ed37916cc584c50c62e9ed88bd817510031e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"b34c5f172b4f093499c58ee6030b690def2fd9a023cc04bbd51cc4aa4357c238","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"a56b4b74752340586b385b92d187dc7f91ffabec9e818120562130dbb9316c0d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"e4f685f754ce7c792af53670c127b3348a85ae89d65d1e2a585ed17cb9d57f19","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"4437c1999aff1f0f0cca4047b249c217cc5793609bbdd8faaf1c4b56e96fc139","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e9fbfdda445c85bc29a7e07cf54f9d670b372862299df304b0c594ed6ad7940f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"f79f144c58fe504441d2922b73de78e1ca93f167f7c22584ae526e5cc9195caa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"090e8b11a6450a30567413a77e025efc409f35d42416fa4a27d3540fbbc32710","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"dfe0923f048be2097d89683fae65dd5e84bd9844a6e1aec51a4ae68520e7f1af","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"99ed1b3d33182af3bf2d8425f501745467dd902e3f7757d15f06a25b448ada41","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"486c1ef641c62a561d549b6acc101ee61feb772ccab8b0017dba9838ae588165","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"e11921e1761ceb8273ee0aa4190ab30b87cb1228fbd7a534ac0545e6f3611eb9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"e6d564ea40761c4bc927e64dcda1b2285b69351f018a7b13face5cc5b1625bed","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"92990ba3caf606e0a8d9152e871456d7d5be23ba54dbdd5941f23cfaa1bde75c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"b144ce86c2734467750cd17118c893422298d5843ab6edafc7d0d5f9d6ee0c2f","bytes":5254}},"requestSha256":"700e094dfe124a58cb9c36453f4687d1baa4308998b62f46beb129db589ffa0d","metricsSha256":"77d5ffa81bfdd765f328fe66ed9df3b675034fd28091c0bdb129b8e1d7fdcba5","continuationSha256":"9c7d051191c5675e1e42b76c4c86e9d5c9db2ebac8fedaba94e08fc0eda64e91","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"e9fbfdda445c85bc29a7e07cf54f9d670b372862299df304b0c594ed6ad7940f","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"f79f144c58fe504441d2922b73de78e1ca93f167f7c22584ae526e5cc9195caa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"090e8b11a6450a30567413a77e025efc409f35d42416fa4a27d3540fbbc32710","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"dfe0923f048be2097d89683fae65dd5e84bd9844a6e1aec51a4ae68520e7f1af","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"99ed1b3d33182af3bf2d8425f501745467dd902e3f7757d15f06a25b448ada41","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"486c1ef641c62a561d549b6acc101ee61feb772ccab8b0017dba9838ae588165","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"e11921e1761ceb8273ee0aa4190ab30b87cb1228fbd7a534ac0545e6f3611eb9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"e6d564ea40761c4bc927e64dcda1b2285b69351f018a7b13face5cc5b1625bed","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"92990ba3caf606e0a8d9152e871456d7d5be23ba54dbdd5941f23cfaa1bde75c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"c92a7aa08bae6d325ee04c1311d195dabca843ab773ee46a9799ce809c2e6752","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f485ff21655ebc865c517f4b7ebe593a9044daac83058cb3b7c6c027c365140e","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"66124a8942fa1e4b467eb3132973809dbefde0a4f93e6399462369a9a8c3dcc8","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2482115","recipe_identity":"e92e02e398e7ca62a11f30e1d904a947e893219e583a2316e7da4f1863bf7774","loss":3.451563449735407,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"3f4802d68b7e643ad20d947fe2ffbb1ebb7ec6761d7f795f17663ccc121e5375","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"5d4872b064d34f2a423a00207b95ef2a10f5a4d4d3a27d72abba5f4d3b3c0e0a","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"7a93e6428976facbbb91b99c74f8a20e053fe66f6f07771467afc0c419a9e2f5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"889e0ae5bc8e8ed1cf28f63edd2a110236918169cbb8b746b2a3f24e3efdf19c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"ae3d4b75948d8c845204bbf85e87c09cbc907a5db0befb28a78d187ede3db31f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"f4033d01662e2de787b98f7079421d4e69e05c6244ef3a9117c7adcde8a3ca80","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"4f66cde24c42588a4d48ca67ca63a203951b01e5c9ad39a8e5024ba03e2578a6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"889e0ae5bc8e8ed1cf28f63edd2a110236918169cbb8b746b2a3f24e3efdf19c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"ae3d4b75948d8c845204bbf85e87c09cbc907a5db0befb28a78d187ede3db31f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"f4033d01662e2de787b98f7079421d4e69e05c6244ef3a9117c7adcde8a3ca80","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"4f66cde24c42588a4d48ca67ca63a203951b01e5c9ad39a8e5024ba03e2578a6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"e3377a7b418694e39234eef6e9050c72fbef6a4be2b9ed3d5bf4914edfb6b3e0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"c578949c2c7bbd493686b33e21bbe0177c6483d265be12d3bdffc6f30e06c9f5","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"0a64d40c3094b869eba9ce9adc406ecf056a5a29585e00539dcc791ec106ee87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"fd18bf50d0e0a1c4e8763aa5c391a78e8b8310287580da6359f6c948224e3f0c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"cb78b791c6e3890d5a2ed1ebe9b14235de63b4c0889a0cb2e2d0ab9718928e47","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"395ec227b5aedfc59be1b91e62a92715d6cea94f5e446a94e8a7c4b12b3482e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"917fa671b6b9ea348919bb89de18d190b2a209ad0ed2c56e31dac793d3876405","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"75ded4be692f859fe3b6c96c3ef461edb4bc89f3a3fea60393756c3481164c2b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"300552b8a5f8e87533ce73dd831979931e59bc0a739543335bd4002c0030ac4e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"e5dc0e0c60fb0fa57067b1c41606f823a689fbff49245c7072722dec96e545b3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"0a5a287258a876ba88b6a1cdbdc6a7e1517ec58899a1496e3ae88251e2b51201","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"7e4718040db83d13a497a08d70b4fe572507401a363d734b8ab3c80fca0e87a3","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"2090583b6a388fdb64f3bb0967ca6d5a3f6c8f0eaaec6bd8220eeb1d47b1c950","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"3f4b453865119a12c11ddd2565d31703f8ac4ae427def0da517ae740dc6a8ac0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"1f4cefe0d2b39eec066f3625ee5a86aab37863a6f52f399ffb28f017a5114905","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"89312a784ce1ed9b7046da1f6c6a5ff604e6e413d755d711b0c8b767ee96f202","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"7f14408f427d3b9449fc0039a722afd083aaa6897720a5d9cd6322038f45b290","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"74adca4253d0f045fbf343d198237537230ebbbfe3bb54d14710953f5b5765e7","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"1630b278a835d8d5d68f29baf84fc5be75f3e3c2906aba955902fa7b3f903324","bytes":5311}},"requestSha256":"76157ecb99fd7863f9488ebc3c7b48edda4927586b8215d416524205466149bd","metricsSha256":"ef26a1784627fcf9611d70b13899ab3243be05415da47afc0f5e4f266bc28911","continuationSha256":"95b828c07a43c33bbf5750e7c93a65ee1e8d506049257077f5260cfcfe359602","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"e5dc0e0c60fb0fa57067b1c41606f823a689fbff49245c7072722dec96e545b3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"0a5a287258a876ba88b6a1cdbdc6a7e1517ec58899a1496e3ae88251e2b51201","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"7e4718040db83d13a497a08d70b4fe572507401a363d734b8ab3c80fca0e87a3","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"2090583b6a388fdb64f3bb0967ca6d5a3f6c8f0eaaec6bd8220eeb1d47b1c950","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"3f4b453865119a12c11ddd2565d31703f8ac4ae427def0da517ae740dc6a8ac0","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"1f4cefe0d2b39eec066f3625ee5a86aab37863a6f52f399ffb28f017a5114905","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"89312a784ce1ed9b7046da1f6c6a5ff604e6e413d755d711b0c8b767ee96f202","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"7f14408f427d3b9449fc0039a722afd083aaa6897720a5d9cd6322038f45b290","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"74adca4253d0f045fbf343d198237537230ebbbfe3bb54d14710953f5b5765e7","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"2376e089b88f80f7e9d408c8d95e1ee8626ca5bb3422a6d8b0c66636e59a290f","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"59c1f97dec45d437e3ced0d08d7e9cf10d557010dd900ec9a422a82df2ee5e7c","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e92e02e398e7ca62a11f30e1d904a947e893219e583a2316e7da4f1863bf7774","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2482116","recipe_identity":"3e3b7bef8628c4bb881c7c66752c238287669d6a9ed63e9f2c20c5cca951b854","loss":3.454236601633877,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"6e09dbabec2855a3f12f50b279849690ccb5f40e4b0dbad7817183b94c533ebd","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"74f500108761d13cdb748c9c9c88951ff60fbcd74e3ec5724200ccd99875ec9e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"3ad4e6c842a4571e3ceb93d930b6d5d95238e44ba4ab7502fb14907f24141053","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"e5741a10b9218dcc9f272c75f9e46547d25a89d7d4dc0c7b4c9f24fee927fbf3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"f8010726309577881aec52ef4365aa8c2d5dd9f2a51701df8db9280bd8f72479","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"2cc0e629011b88000a72ed0d2803e229559e3e81dde8fd3a3e3c89bda776b3f5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"59a1b049be836a40f1ff3a43f57f2367d379e5e0e9a1d9c978174c33cad47177","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"e5741a10b9218dcc9f272c75f9e46547d25a89d7d4dc0c7b4c9f24fee927fbf3","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"f8010726309577881aec52ef4365aa8c2d5dd9f2a51701df8db9280bd8f72479","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"2cc0e629011b88000a72ed0d2803e229559e3e81dde8fd3a3e3c89bda776b3f5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"59a1b049be836a40f1ff3a43f57f2367d379e5e0e9a1d9c978174c33cad47177","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"b2765b30e10868d3c7596a6801fe7a00b4f511b7709f8c2affc68f22daf97b1a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"fc1291c5bf6b1e37571293edb721d396bca71eaa9eda61962482885d253b3817","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"e123f49d1cbcfae8228c2b0aa143c9bf92b2c4d6aef650eb41ff5da09620b983","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"21cfb260bddb6a46e9db464f50210ee1ad930ab944468e73ebaf572f8c1947ed","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"aac3d1db910179fd173b467169954912bbf9eba5e07324248c547f8c6cc16e81","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"a9966108ab87b16052b9ebba1685ed3800dd5495235753493782426f4bdc9f8d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"a1febc4ed22864fa36bd914bb9ab14060889f128c9a0da1f6bdc49443ec2fc68","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"9e27b110a373269808ef699246341d13c413baa13cf8abe9be4f8465f1905e64","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"0068e4670879500f96002e7d8d50aa5a6625b718139b02a0af18b92aa1203390","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"4b72c393bdb9f69e7bdce3782ec2e64227faa1b0ee3de0d0d4a4f803f6207b6e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"2e757de9d41a416ae46599d41a01383ebd3a78d232831f71421eb7eb62b480b3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"a12c5ede4bb4339c8fda972c67ae0d0fdf4250ba5182ab1fce57b6cd2d61300d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"ba09000ca0408ededf8fb56a253129b7acd6be1790e127d3eb30bc7bb26aa0bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"1eb018df4adedcf5bfbf7861260903e3134799ca77c8842ac9db0902c8ab77ed","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"3eaa3dfca30e14dd2ee830f0da8f4d50b86cc18a56efcf339066fd6cb41fe1c1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"92e36851088be6c13e948759e3d52ed542199406afd6be7f3dea35327d745fad","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"0d0aae86f9c694b7c25c1d92a235fdecd81289297596cf7afbf162dd573a6685","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"96f94db00f0a52e444a9356e3b62c235ec7d0a9cb88c3803282b322c4378dd4e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"5ee253915cc3d7136a9b729a0c0a858fc5b3d931b03414638d58353ce2ff1677","bytes":5340}},"requestSha256":"e2d420a73265914538f95fcd9d3a1de6031021b1dcea179cb92e76ab0b15fe7c","metricsSha256":"d135172c9b575f8056bd95678fa9d9a684d35ce3a2aabf9a9e38551e523260a7","continuationSha256":"1a4f5e14d51778afe0705a578c72ea444620e4d776f6d77790583fb3eb1f3c76","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"4b72c393bdb9f69e7bdce3782ec2e64227faa1b0ee3de0d0d4a4f803f6207b6e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"2e757de9d41a416ae46599d41a01383ebd3a78d232831f71421eb7eb62b480b3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"a12c5ede4bb4339c8fda972c67ae0d0fdf4250ba5182ab1fce57b6cd2d61300d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"ba09000ca0408ededf8fb56a253129b7acd6be1790e127d3eb30bc7bb26aa0bf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"1eb018df4adedcf5bfbf7861260903e3134799ca77c8842ac9db0902c8ab77ed","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"3eaa3dfca30e14dd2ee830f0da8f4d50b86cc18a56efcf339066fd6cb41fe1c1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"92e36851088be6c13e948759e3d52ed542199406afd6be7f3dea35327d745fad","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"0d0aae86f9c694b7c25c1d92a235fdecd81289297596cf7afbf162dd573a6685","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"96f94db00f0a52e444a9356e3b62c235ec7d0a9cb88c3803282b322c4378dd4e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"f21d93a66f0f0fd986d152aa23c7ba949e6c0f8bdc0718858228bbab2566bf4d","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dd7156a6509ab9eb2075a92ce9f55e748dae4ff9b97b82887a505386b955e910","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3e3b7bef8628c4bb881c7c66752c238287669d6a9ed63e9f2c20c5cca951b854","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2482117","recipe_identity":"6028f48984e92aa1337ed658409ab8fd39c48128b96aaf69adc8615b99a9d14d","loss":3.448060505913961,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"6eddaba511c1526fa272188b868ac859a46972ee847247bd2ec6dff85aa736d7","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"ee334d1261e05faa3339899d600163e4fee359b1542058f54af1d0eedc7deb3f","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"6dcf9de7346a6f2689120ad67c5b2e4372ca3f2bb7504df8b7ced362e90587a6","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"250e18109c2c51332b2da730616b945dd5ada220276e90981ef87488d16976f8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"740877f1920c97415cd76cc017e0901246d9b3f94b90fefc18e020c37499f467","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"e195e98ed40dfbe3b72bf86e2a498b6007abc30e89ce8a8e8d675ee74ae01f46","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"4d5c94275349c5c613f5bbed2fd2ec44eda6c04323cad5adb95158d370528640","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"250e18109c2c51332b2da730616b945dd5ada220276e90981ef87488d16976f8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"740877f1920c97415cd76cc017e0901246d9b3f94b90fefc18e020c37499f467","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"e195e98ed40dfbe3b72bf86e2a498b6007abc30e89ce8a8e8d675ee74ae01f46","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"4d5c94275349c5c613f5bbed2fd2ec44eda6c04323cad5adb95158d370528640","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"5409eeb6cbadc55c7df8e153ac76f4037f99b74756949003b6793520a5478f51","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"ac8ef820500e8999cddad8c7e5e4d69cabae0951f29ebeaf868f03fc93e6558f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"0fe2003a43d7b6e57260231ea3351120e8a15a4f77e50250f289269535b657a9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"deb75348ec677ee3f7397474ceea22c9c6735c1f040ac2e6ab57f32d6fb72ddc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"bb384dcb3f3448a798d543e81910053f6cb33a01273f4db5450abf9d8b7f9d6d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"d8a1bac51772fb14d968fe412b90750a59f46d689a03d3f32085ab6cc37b4310","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"3cf480cc859dd3534cc2000ceb5fcafd837e782094e72324e42366ab54d907da","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"e1ba070519d4c22aa6e945dbec015503f22decd7c883f98745cb8b5c14672fae","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"a1037af21826a2823eb00dc5c435bd05f2a32f6ce07e5045bab4f649e9019ca9","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"827102717adf6c28c522b81749322650d67ec6a69b042bbede76dc6bcf8e733a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"a045304b455f348a68cf677f697abc370a9ab73fd2cf21f434719c9ffbb99f6c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"5d354bc822924b54144036e0dbbde1b7b6002757b77d88317f9c7e9f97bf5b11","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"d213f0655e7c35c35624bd8b87fd087ed83861c7c9e9db360cfd8332c6399428","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"a99bd471846ec06edd5346ab2797a1f2511d008267b50b1ced021fd3e5447ab6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"76cf7d4cfbb38252a3e67948c98045cb4385ebee6862d160bd567009576beccc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"d509c1f3970477ddde1da80e60f0b3466989e39cf523159c3e3107e26bcc4931","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"9803771e681df8b447df584b9f2ebff8d6acd0e7b66968602ec39cec7508cdef","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"55507e4abbc089c202ce26f6e1536967cbc1ac3e6b7d8eca0c07502c0164df1d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"d0d59d1b05fa6840c8db73763aab6f1c7d4a833ed6d39343526f9e9107dabd2a","bytes":5315}},"requestSha256":"cbb4256c1b809941bfa2986af555efb07c592833a86406e3c334935d8814d3db","metricsSha256":"95d6691f0666df169193993bf87a9b7672b413a7e14b10baffa9b9ee8c366bdd","continuationSha256":"71f1e8c8a289e896a57f63270d9fb2d51fb302da62b4eb1fadb3a84b54945d37","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"827102717adf6c28c522b81749322650d67ec6a69b042bbede76dc6bcf8e733a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"a045304b455f348a68cf677f697abc370a9ab73fd2cf21f434719c9ffbb99f6c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"5d354bc822924b54144036e0dbbde1b7b6002757b77d88317f9c7e9f97bf5b11","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"d213f0655e7c35c35624bd8b87fd087ed83861c7c9e9db360cfd8332c6399428","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"a99bd471846ec06edd5346ab2797a1f2511d008267b50b1ced021fd3e5447ab6","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"76cf7d4cfbb38252a3e67948c98045cb4385ebee6862d160bd567009576beccc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"d509c1f3970477ddde1da80e60f0b3466989e39cf523159c3e3107e26bcc4931","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"9803771e681df8b447df584b9f2ebff8d6acd0e7b66968602ec39cec7508cdef","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"55507e4abbc089c202ce26f6e1536967cbc1ac3e6b7d8eca0c07502c0164df1d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"e39fa898d2b5de93cfcd14c003431a0ad8fa024756fa977fe7ef8bda7f76b416","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ea414fb12070c56c8608295408e865169a0d8d1a27426f7d95c0a26c3a68c350","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6028f48984e92aa1337ed658409ab8fd39c48128b96aaf69adc8615b99a9d14d","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s42","horizon":80,"attempt":1,"job_id":"2482118","recipe_identity":"71e913d641d991d6835a22dc1b9ae2f7ff786a4b9f675e3fe1ea1b79208257fa","loss":3.394842061696633,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/result.json","sha256":"87e7ff65b0a2443a074c63cb84b9760548512d84bc9225703c2d27b524eb7c1d","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.pt","sha256":"c719c0485978286e8631adf85048e072844a2d563861a66cd1421cf15c562a1c","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/final.safetensors","sha256":"f786193d6e9dba2183cce9068d4adc9c31f2d41ae6f4b1ebf4b10b85af1eb1ae","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/final.safetensors","sha256":"76ceb4159905c55693a79afaa996976d4772b2ff2b5ffae6190e69ddc6792205","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/final.safetensors","sha256":"64e7accfb2bb89af2b3620ef1168bd0dd18e042083251d83f05e5ae1399a67f5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/final.safetensors","sha256":"89847f7db847c73e205d0402b2bf15c3c3dfac2b421912d5bfa7fe5bfa926696","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/final.safetensors","sha256":"69c2757125fff3fe6dbdf902cf3acc0ada0036b5a11b389d95666193da459083","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/final.safetensors","sha256":"76ceb4159905c55693a79afaa996976d4772b2ff2b5ffae6190e69ddc6792205","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/final.safetensors","sha256":"64e7accfb2bb89af2b3620ef1168bd0dd18e042083251d83f05e5ae1399a67f5","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/final.safetensors","sha256":"89847f7db847c73e205d0402b2bf15c3c3dfac2b421912d5bfa7fe5bfa926696","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/final.safetensors","sha256":"69c2757125fff3fe6dbdf902cf3acc0ada0036b5a11b389d95666193da459083","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-143.pt","sha256":"77b92c4c37620112e4a02e4e63b4771b4f9250904edab335c938bbf401f41e2e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-143.pt","sha256":"e95abb85ddf10c064580b711acbfc6bd8fe041baa67ddea6529e2adee67ad0d1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-143.pt","sha256":"72d03b65fc62d0f6437b7adea0e84fb4d18e3e470cd7b492dc15babae06f4a01","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-143.pt","sha256":"9eafea003cda8539d37835fefe323357482b1559d4eb691d8269492f1ece0ccf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-143.pt","sha256":"2049c2f4b9fc3cc0f6855f184d97b6c5c4bc5278798477f16a764bcf609e6669","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-143.pt","sha256":"659a171504c1c74ae2ad8ca9cdf97cada1ba0e37965d954742c6af5a063c0419","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-143.pt","sha256":"87b23419936dfe5261c6646e3e5861e4025545a26efe6b9c2d0a9e30266c5c34","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-143.pt","sha256":"d7bb029b0dd56be8d467a89cbbff7363c0508eb8c2729661577d839055a9174a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-143.pt","sha256":"266e62972552678565b45b8d7f7aaf62edde4cb3f8fb6f7435da9c0b444b331e","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"0ff4bae7a842f35c0a185fc93c8764aec5f540e0dc1078aafcec28af6b47b07e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"e0dbe30f061030a304d36ee4bb786c434169a66a80cfb2fad9aa1dfbb8cd0b85","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"2f3fc1b9c3114c138a13cd23c3fcc48efef08215877789a1c55dcd9f00f590dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"465f065871b7ff354fe7b2d7dd979600d1d22c90cdc34206edeb57b2ec571a08","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"1743598b435d9565ec003cbc9633f31cdaa99614c57733024a20afdd4fc004ea","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"e2446846adf5b1afae061885779b994ff74f78e7b727767cf0325b963e0c791c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"8cabe4fdf6bcfe03f6ef32b892ff52bee7f0af3759f4fd38f06f3701d17cc93a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"f4317b2d34c247af2681ba44ea1025b54486b18c53e92f1f7333017c06bc3b94","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"c2af7a0fe016db28113ca7f018f0d0068e627ec6a21e049aaed2d86092265296","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/environment-resume.json","sha256":"e1b86dc9e2cc1837b792cca7cadfe4d85f6f0f93c37de156a8439fb9ff999609","bytes":5257}},"requestSha256":"ee958b9dfab114248c9add59103bb55922e60f6b79d418eb4896fb9895047d32","metricsSha256":"8f2048c957e1b722539e97d3efce639ac723d10f39713dcf931b7807504c1b9d","continuationSha256":"fc36db680614ac72ec574b219d75f0d1fa5777424c31bb3567bfbc045e1768d3","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/epoch-072.pt","sha256":"0ff4bae7a842f35c0a185fc93c8764aec5f540e0dc1078aafcec28af6b47b07e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-000/epoch-072.pt","sha256":"e0dbe30f061030a304d36ee4bb786c434169a66a80cfb2fad9aa1dfbb8cd0b85","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-001/epoch-072.pt","sha256":"2f3fc1b9c3114c138a13cd23c3fcc48efef08215877789a1c55dcd9f00f590dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-002/epoch-072.pt","sha256":"465f065871b7ff354fe7b2d7dd979600d1d22c90cdc34206edeb57b2ec571a08","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-003/epoch-072.pt","sha256":"1743598b435d9565ec003cbc9633f31cdaa99614c57733024a20afdd4fc004ea","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-004/epoch-072.pt","sha256":"e2446846adf5b1afae061885779b994ff74f78e7b727767cf0325b963e0c791c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-005/epoch-072.pt","sha256":"8cabe4fdf6bcfe03f6ef32b892ff52bee7f0af3759f4fd38f06f3701d17cc93a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-006/epoch-072.pt","sha256":"f4317b2d34c247af2681ba44ea1025b54486b18c53e92f1f7333017c06bc3b94","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/node-007/epoch-072.pt","sha256":"c2af7a0fe016db28113ca7f018f0d0068e627ec6a21e049aaed2d86092265296","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001/continuation.pt","sha256":"6f68628bbdf3ce974ab41ef011f2d021011c5ddfb25fbaafa112217665d82707","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e6f2593db657136a1a8e50acaef613c801fa83625316613394a80c83980e1c4d","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"71e913d641d991d6835a22dc1b9ae2f7ff786a4b9f675e3fe1ea1b79208257fa","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s43","horizon":80,"attempt":1,"job_id":"2482119","recipe_identity":"4f4f86ec8317e2dab01fea347fa5db4aa305b71e290bad98391d07e831d2db0a","loss":3.3995739058385386,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/result.json","sha256":"5ca45bed439ad3d22f68d513c9cfeafe951892d402815c5592ed4983285c1c43","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.pt","sha256":"7c35ecda7bcf9bcbba9321432a85e99d3232e5bd4506e01d22de3ed0aabc66b6","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/final.safetensors","sha256":"a2bb1346b7ee4d4d2718e89b85cc7a4d74431fe2abacc2de0a876f0a7279b3eb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/final.safetensors","sha256":"eaabe0b303f3944a817ddab99bf2d541f9698669213a9a3deb0a3e9a70e13645","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/final.safetensors","sha256":"a26b039856cdc3b49b1db6af17787eebc03bbe842047a5016f4aa7043248f0bb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/final.safetensors","sha256":"35c8649e63e4156ed1d9aba9e97f4839a7f24b54a7d1159c6baa02b4e9a61c7e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/final.safetensors","sha256":"d55c32c219650d09edd83eea6601584ae8e97cb44d5b2c252a7e30fa9fc538f5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/final.safetensors","sha256":"eaabe0b303f3944a817ddab99bf2d541f9698669213a9a3deb0a3e9a70e13645","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/final.safetensors","sha256":"a26b039856cdc3b49b1db6af17787eebc03bbe842047a5016f4aa7043248f0bb","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/final.safetensors","sha256":"35c8649e63e4156ed1d9aba9e97f4839a7f24b54a7d1159c6baa02b4e9a61c7e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/final.safetensors","sha256":"d55c32c219650d09edd83eea6601584ae8e97cb44d5b2c252a7e30fa9fc538f5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-143.pt","sha256":"723bd1c89674edbf21afda9a7a7b07396548a42f88a594aceb2489a5f3fb20ca","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-143.pt","sha256":"432fa17566cbbcc2e35e86e9ff721b1d8705a0e4cd84ad51570c33b58ae53b39","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-143.pt","sha256":"8a132ad06a69bcc27f636c8752686356a700b2230b1bdd024806a1040b936c38","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-143.pt","sha256":"186c0b436471fdb19e34e349387346254f216167182521afbb3b1c9a19bcd5dc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-143.pt","sha256":"1eab98f0320a5d3f0f4b445586d1450e4cdde24b3d37df8b8322c5ddd24d5ee9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-143.pt","sha256":"54c90bcba8048692fe51d4a485da712910ea2bf2c03c3f48111b8c8a2807c4b0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-143.pt","sha256":"b39d144b99a0d506e3afa0e0017537a2a3cf6e119105eada0af72af0a1ddaa86","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-143.pt","sha256":"a9e40517551c97e640752aa36458e9e6b8c56b2901efe601fc89c54b040bf597","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-143.pt","sha256":"8a45d7ce3f26325e63320b43b4ff382b04f5b5598a37f2f5f9a7927022a50d5d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"930f7093adc05f6a935acdd593262e95895cb2190b2262feafb9dec1451fe5a7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"33ecff47fe5e88c510b98e0e084e18eb1f5afd51e34e9cae972b7f8ba2644c9f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"2de344c73bbb43f64ce835b047a024c5fdae19ca7d4e6f1950bd9fd7018a886b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"5ccdf7d97059fc8770cb9254a3b5ef0cdc5433f4329442bd532975fbfc0372c9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"2c2012f6987bd8d59525d00949c7d23f1cc2a842203e6abcd8b3052ccf92ade1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"17aa3b5880035ad3ef0dbe0284d18cdfdd8d7df4a5aaeda2aefaa7cc08258bc7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"16f53742e3265ce08577beee349f621c20bc8008069432ca3be009ff43f375c1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"d8b4c96133c468f07d1a1016b66e8910ab9fe66f105dc8995428a198da8e9af5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"d864b740f9835ce6ca025fe4dfcd9810a15ac277fdd4b5ed3f04f2352c76c03d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/environment-resume.json","sha256":"5cf341bbd2c0181dc142b12773602f5928230579254f29e749e42f43dadbbb1d","bytes":5341}},"requestSha256":"48a78750894cc5d1fcb336d36b0e32529f5a4db5b14e9652049ce1ec629b8cdb","metricsSha256":"cddd05469c749da0fe7a33950d64266661b157e80b46c126d4bfc5bbfa6f5f3e","continuationSha256":"99ab80657d9331e3e613168e7e6c10dc37a071c4d3a10a306c1f976d502ee3d9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/epoch-072.pt","sha256":"930f7093adc05f6a935acdd593262e95895cb2190b2262feafb9dec1451fe5a7","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-000/epoch-072.pt","sha256":"33ecff47fe5e88c510b98e0e084e18eb1f5afd51e34e9cae972b7f8ba2644c9f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-001/epoch-072.pt","sha256":"2de344c73bbb43f64ce835b047a024c5fdae19ca7d4e6f1950bd9fd7018a886b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-002/epoch-072.pt","sha256":"5ccdf7d97059fc8770cb9254a3b5ef0cdc5433f4329442bd532975fbfc0372c9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-003/epoch-072.pt","sha256":"2c2012f6987bd8d59525d00949c7d23f1cc2a842203e6abcd8b3052ccf92ade1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-004/epoch-072.pt","sha256":"17aa3b5880035ad3ef0dbe0284d18cdfdd8d7df4a5aaeda2aefaa7cc08258bc7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-005/epoch-072.pt","sha256":"16f53742e3265ce08577beee349f621c20bc8008069432ca3be009ff43f375c1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-006/epoch-072.pt","sha256":"d8b4c96133c468f07d1a1016b66e8910ab9fe66f105dc8995428a198da8e9af5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/node-007/epoch-072.pt","sha256":"d864b740f9835ce6ca025fe4dfcd9810a15ac277fdd4b5ed3f04f2352c76c03d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001/continuation.pt","sha256":"6b5e678a5341f8992fab5e291c5f11829394c7ab912f0cbaec57212948d13cc4","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"43dae8aafe05e36d73781226b5d94144d1b899d74dcf461905204cc1f8cf9de6","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4f4f86ec8317e2dab01fea347fa5db4aa305b71e290bad98391d07e831d2db0a","loader.blocks":1593984}}},"awc8-wsd-h80--0.006--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.006-s44","horizon":80,"attempt":1,"job_id":"2482120","recipe_identity":"0b0bcdd1a071a39307e226d32a320e0e37de8d87d37fed63ec0614411405cdc0","loss":3.3955923765626985,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/result.json","sha256":"593808e62775995cdbb8469be119cdaa300daa5f1b75c7c3ae4c4e90e8ebe966","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.pt","sha256":"c3173a320f8e67d55542c71417b289b2358e1403f8dab4cb880652edf0dca6f7","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/final.safetensors","sha256":"9fcff56eed0ec018ec2f206ec5ed31b65d7823592ad075caffb7c1c4df0fd317","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/final.safetensors","sha256":"bdb77583cf2a8d1b768cba6cfb329861e63f7c74023f805674247eee404310ea","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/final.safetensors","sha256":"b8f4b4305291138afa41cfdb6d6c7d83e7519d9ce1e53ddf6d078fa48ec8524e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/final.safetensors","sha256":"f902887fa3a9c0a7095a7d2bb343b63474714e934d7e11a9e0954f778e3ecd02","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/final.safetensors","sha256":"8f0efedfe736f0d3f7a2b310f8cb7be4849b74e4a16ddaa932e008951c700799","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/final.safetensors","sha256":"bdb77583cf2a8d1b768cba6cfb329861e63f7c74023f805674247eee404310ea","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/final.safetensors","sha256":"b8f4b4305291138afa41cfdb6d6c7d83e7519d9ce1e53ddf6d078fa48ec8524e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/final.safetensors","sha256":"f902887fa3a9c0a7095a7d2bb343b63474714e934d7e11a9e0954f778e3ecd02","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/final.safetensors","sha256":"8f0efedfe736f0d3f7a2b310f8cb7be4849b74e4a16ddaa932e008951c700799","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-143.pt","sha256":"166b5f77d2f2000f97f6a1386cdb735a2c40bbac246eebe630f7ecc9149aaf39","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-143.pt","sha256":"322fec901480292ab3d640f86461578e827efe25cdf5e71828be57d31c2135c6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-143.pt","sha256":"fed815de1c005f91d5fe7533bf431074fc7c9103bc5e90296b70df05fd97005b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-143.pt","sha256":"0005479d416ec8cf7bc7bd8cc83b94100de8f7be08017531d353e2b70a982efe","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-143.pt","sha256":"dc41d547bd6fa24347243a3042af5a1cf80b84126f94f1ebb64db2e475d58300","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-143.pt","sha256":"352bd45cb81889e1e58dfc0a9e3fdadacce148e1053bfe6d948c8deb8954fd6c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-143.pt","sha256":"86760b0b6c6467dc28d2d2989fd42b41232940874e8e6f3cf1a08f8232aed510","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-143.pt","sha256":"9b40310f3c070e20d1b701c9045e47ba81a4e914530f85c6d020b0d59fcd19cb","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-143.pt","sha256":"b603c4b997bb465d4d8a322fdc7903482a20fa9a91f08151e26d6069bf1bee66","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"724fdb8dd67c02cf1abda03f4dc09418ed5801f1b1c5fb2869e0d39e1887a795","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"41c44d0df20eea437e851fe67d949b84b6c6888898b79236a835bf3ed6d92d3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"f9a08c4cc932f3bb30a9266955c12cef9f8cb35ec761332a4ecc558426a55167","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"fa4c467be4aef10ae48954db67a362acc5df52bbda2dad5829309699c89dcd36","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"e05238fe1c6f6adf84e8fa5a52f7b3106ee2746d5b880f408557ebf5f6b68888","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"c4b0616483ac2a6f7a574f2145f5f8b6f95b53b1cd935b20a3478e5c91f07d67","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"087333f504f89424fba08f214d0fc36e26c9d55111d1403c158c6fa69df61b2f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"f32816ad635cf020be3a3a2767c56b5996caa3db6f4ab8b326863c97900438e9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"2c4061505f25db7f891a2edf004dc63e3ce4fc993e808f23dbd90ff699fea811","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/environment-resume.json","sha256":"341c844995086715d3c85b594e3e14edccd88989699375987be0a0eb16a25de3","bytes":5311}},"requestSha256":"530f2d79cdf9ece0cc196639465612121700fc6129c97876b0a2e43488ac7677","metricsSha256":"3e93ffea7ade19762b619c73b52cdf7541dab10eb4ba193884c6e92d92b64d54","continuationSha256":"1660772d0ccb4b1f86868aff126e6fe3fe87095f832bb08bb2db98b9b432db28","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/epoch-072.pt","sha256":"724fdb8dd67c02cf1abda03f4dc09418ed5801f1b1c5fb2869e0d39e1887a795","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-000/epoch-072.pt","sha256":"41c44d0df20eea437e851fe67d949b84b6c6888898b79236a835bf3ed6d92d3b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-001/epoch-072.pt","sha256":"f9a08c4cc932f3bb30a9266955c12cef9f8cb35ec761332a4ecc558426a55167","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-002/epoch-072.pt","sha256":"fa4c467be4aef10ae48954db67a362acc5df52bbda2dad5829309699c89dcd36","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-003/epoch-072.pt","sha256":"e05238fe1c6f6adf84e8fa5a52f7b3106ee2746d5b880f408557ebf5f6b68888","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-004/epoch-072.pt","sha256":"c4b0616483ac2a6f7a574f2145f5f8b6f95b53b1cd935b20a3478e5c91f07d67","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-005/epoch-072.pt","sha256":"087333f504f89424fba08f214d0fc36e26c9d55111d1403c158c6fa69df61b2f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-006/epoch-072.pt","sha256":"f32816ad635cf020be3a3a2767c56b5996caa3db6f4ab8b326863c97900438e9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/node-007/epoch-072.pt","sha256":"2c4061505f25db7f891a2edf004dc63e3ce4fc993e808f23dbd90ff699fea811","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001/continuation.pt","sha256":"a252addb60566a99e615041c86668f762d312dc3412ff791e333ca954d200710","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3db56bb253af1abee8ac4f594bb4cc7934f1ea50afd9ee571e1a13d8e7f4ba2e","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.006,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.006-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0b0bcdd1a071a39307e226d32a320e0e37de8d87d37fed63ec0614411405cdc0","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s42","horizon":80,"attempt":1,"job_id":"2482121","recipe_identity":"138744e39b0243bc794b66ab5af9da07e9b6de0aeff44552a5de72bf45f87f43","loss":3.389211589365249,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/result.json","sha256":"e3554ce4469cb1e7db888ea72702a0e1ed47381085f1468a49d0b2ce0e9128de","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.pt","sha256":"c884aa3ff2775fa2fb7f8300528b06836f24da6825af8ecad10f3997618b12f8","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/final.safetensors","sha256":"ca07143bc103955a8f4c8a77a4768303e8f465e1c55455ad5ca72259b04d88ed","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/final.safetensors","sha256":"454d8ea1e7b3303d9bbf327a20c6f90879ca2a4e4f7c14d49ba092017acf9bd8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/final.safetensors","sha256":"d39a5684534d758abb112c618f2c10b8d2d2b983543ce1e5a64d5e5db292fe4f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/final.safetensors","sha256":"e032b010010894052bbc99074e84b3cc8b83d33062a8393f88150025ba599581","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/final.safetensors","sha256":"9ea26c83cb321bc51bbdd51de208c19a583ffa2221233692d964443c3a069a96","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/final.safetensors","sha256":"454d8ea1e7b3303d9bbf327a20c6f90879ca2a4e4f7c14d49ba092017acf9bd8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/final.safetensors","sha256":"d39a5684534d758abb112c618f2c10b8d2d2b983543ce1e5a64d5e5db292fe4f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/final.safetensors","sha256":"e032b010010894052bbc99074e84b3cc8b83d33062a8393f88150025ba599581","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/final.safetensors","sha256":"9ea26c83cb321bc51bbdd51de208c19a583ffa2221233692d964443c3a069a96","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-143.pt","sha256":"1f7d983ce35496c29e09a63e5fb1fd7ea9f5b5d73a0fc56605f9e728b8610dae","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-143.pt","sha256":"fd2cafa82c72f1097a335725d218d662bfc7057deb52f5f4ac86e35e310e6e8c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-143.pt","sha256":"7b136246a3b0f1f225ae886187cf06eea9d1e399c533584054170e62373608e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-143.pt","sha256":"3146f1c535447912e8d9665d01efa3f08c109154e8f0f71bb91a89d4c50b2406","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-143.pt","sha256":"77419cba1830ca2c6fa257fbf738662a357b9644d0105f7fde9add67a9262011","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-143.pt","sha256":"f0c9548a75e45e35b4b944aadd9c57f7a3044c1809ff4b74a9437e3874291828","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-143.pt","sha256":"76a0a82dc4663d0a74fd183f1b1f13f39dbf68e63b8e9b925a368d0bb4a4d303","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-143.pt","sha256":"58214329927ae78823d977e8694cbaf467367d54ff71f2096677f0007515ffb6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-143.pt","sha256":"9c3104ca0d2a07e2151a87eb6bf320118dc6675f6b6fbce52fad6928f2694052","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"f8dc276758deab74b2c76dac1e26d3df940121e5ec1128e853014a683562b5d5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"5b79e245474c4a002b0b10b95ecf11ca03f2af2cc588393b4b919197bfd2bd02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"8051426777397fc73cc99574a8e7b671c9170994a0695c154b2985f35b263ed8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"7f0154be5e2a1a542569140798e3fbe6ae6378c992e8a58bbe47252dbede87d4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"b3993d7dec40fbb05f87b44785aa166d36232c95fa22d2c403555ca7af3919fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"2b8c1f47fdc2b16f14e7b4e1eda803158023f4fdbae8a75b8c3acc22e4131db7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"52911fcd662da56ab42486e6960e1f1672432418fdef4303506c27aef31d2714","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"4076eaf99966bb8229403332a5d1fdf7b8254f002a14e708050f4f8eb220e8d4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"4f41e1adb44e1e6490f9d8b1ca8f082fbb073ad81f3d355b334979bc562c59ae","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/environment-resume.json","sha256":"22017e863ad5f17c027b8fa45b2ec7e5d6f47f82c22f34936304982ebaf51a99","bytes":5340}},"requestSha256":"97b3bb66de8bcadda28c968acbe926f4e2040257d053bcb4b267ea0150d91574","metricsSha256":"f74321c8d061a24a36d3cff9bb4a5b8ea19077a24c61dc4356c997dd4c142e5d","continuationSha256":"d3b31256c5784151dd292df7753c54d46cf524f0748d4245e25a99dcad602e08","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/epoch-072.pt","sha256":"f8dc276758deab74b2c76dac1e26d3df940121e5ec1128e853014a683562b5d5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-000/epoch-072.pt","sha256":"5b79e245474c4a002b0b10b95ecf11ca03f2af2cc588393b4b919197bfd2bd02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-001/epoch-072.pt","sha256":"8051426777397fc73cc99574a8e7b671c9170994a0695c154b2985f35b263ed8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-002/epoch-072.pt","sha256":"7f0154be5e2a1a542569140798e3fbe6ae6378c992e8a58bbe47252dbede87d4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-003/epoch-072.pt","sha256":"b3993d7dec40fbb05f87b44785aa166d36232c95fa22d2c403555ca7af3919fd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-004/epoch-072.pt","sha256":"2b8c1f47fdc2b16f14e7b4e1eda803158023f4fdbae8a75b8c3acc22e4131db7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-005/epoch-072.pt","sha256":"52911fcd662da56ab42486e6960e1f1672432418fdef4303506c27aef31d2714","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-006/epoch-072.pt","sha256":"4076eaf99966bb8229403332a5d1fdf7b8254f002a14e708050f4f8eb220e8d4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/node-007/epoch-072.pt","sha256":"4f41e1adb44e1e6490f9d8b1ca8f082fbb073ad81f3d355b334979bc562c59ae","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001/continuation.pt","sha256":"ede56bdcefc6b42d4181555d92d1ce5eea07f08b71d1f830cd4c46d3e384523b","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d1835c98fe6ab1ae8a9112c94dffe867f135802ff1c1bb5f76d1d2a9820d4da0","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"138744e39b0243bc794b66ab5af9da07e9b6de0aeff44552a5de72bf45f87f43","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s43","horizon":80,"attempt":1,"job_id":"2482122","recipe_identity":"1fa2f89a26cbbed64cb297df6248e4ca7da3b1e0652b25e6b43c79c513a591ac","loss":3.3958276538213052,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/result.json","sha256":"76214601aaa3974f2faa091e3f7c68e8e5013a428a3d43f04e2b9fcd3289376b","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.pt","sha256":"16a84a29f6c9c678316a71a0799168fbbf4642a5c360d5918d766acaf2f5c724","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/final.safetensors","sha256":"044e53279de096a4b801cf076cc6d44abdee200d6d70b5668da2da83fbb6a350","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/final.safetensors","sha256":"27bde7dceacb383506190788832e4d165a6433685f1ef3a825fea48f0ec55b9c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/final.safetensors","sha256":"7a2a8fd292d244ecabf0946cbd4215257f3fb00912884ee220ddad28e734af3f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/final.safetensors","sha256":"554968982c8138c29d3c13118bc59d8604d7c56856c5e2ee566aedbc61045d08","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/final.safetensors","sha256":"21767f2993233e8135bc6d192a057c9d5af6cbd5b04e15d746ad27013d3d0154","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/final.safetensors","sha256":"27bde7dceacb383506190788832e4d165a6433685f1ef3a825fea48f0ec55b9c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/final.safetensors","sha256":"7a2a8fd292d244ecabf0946cbd4215257f3fb00912884ee220ddad28e734af3f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/final.safetensors","sha256":"554968982c8138c29d3c13118bc59d8604d7c56856c5e2ee566aedbc61045d08","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/final.safetensors","sha256":"21767f2993233e8135bc6d192a057c9d5af6cbd5b04e15d746ad27013d3d0154","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-143.pt","sha256":"47c80a16db4a42e6569ae1384589e61231ce4a874072a486714430b76a75069e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-143.pt","sha256":"245dde6e149c286688d605c10cd8806aa691aabaccecbe957eec1154057c7ac8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-143.pt","sha256":"5cc949ff00eb45ca98dc79fa0684988ca047c245bcbf8024dfb5e7e91dacfa7c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-143.pt","sha256":"29e0f3a006130b2e4c89da760c229ae34f82975cfeb3944cf14a5f29767e19d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-143.pt","sha256":"d6ef032b7d688d3c94c190360796dd5f816a8f869ffe6ae0d98ce35611554304","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-143.pt","sha256":"670d36f9026b2d3f11ba88d7fa90e8ad12ddf2a390052a2ced35d3bdbd7d1a9f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-143.pt","sha256":"1a231ede8a78e6d5f0e483b18c1dde5c9c571050cda6256c684a25f0a82f6c1f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-143.pt","sha256":"d65c7086b8d07cf6c3f2852eb87875ca88db8857895f70e51dbb9c38f5d4bcc0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-143.pt","sha256":"3eaa5521c48d640dfe07e92cfabb7da2662c9922bef2666b4dc89d843a8c276c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"a2b07bea8609ae0940fac4e88fc6549e9b2addf110f07149ba265496859ad81c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"d40a24233a9e7f20ee7ddb6dc318f37fa381c0c62fb89f84a9e3f88498775837","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"c12ca6e64ee9f4643e4e1cf1480a4452a955ddbbed4a50da588f9009b4a939bf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"3bc347cc3a7dbbfb114630fc0e5ba32e02726eab4b7e2af7b979c145d6d586b5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"cc7d5eebd4f09c90363f4ac16a0739dd8a923c8be1f394a1681e2ec0c3060cca","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"2dca0322b0f7860e36b37e882a1f665fd4bbd63fe3bc8187f74d1c4b7b185b95","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"2836bdabe814c297172b733f2c7d081b32ff132dd5e6d84d372784b7efbef1dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"2bce0298bd13de03d9be4a1f3561c11dee19f907c1483d94590c8ca2ad03be03","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"b1de50eb3a1a7235058497d66ee746b11c9069d96dad3cbf47d4c0e0dd8a8a75","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/environment-resume.json","sha256":"b66b511fb3b56b2e62eaca08c0fe6186941fe2a29dfd990d0556b5b26e753856","bytes":5258}},"requestSha256":"c5bae30e7a8da6498ac3afc7d048b98d6aab9e6f3f6048e07ca170d174ebefc8","metricsSha256":"73475e2d8d72b914e643c60fcd0fc91f03b50772a339bbd1142ce7877eb0d386","continuationSha256":"ea518eb6559bf9e0a139749c13563c78b09e28a29c0ff60b395873f1e36660ad","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/epoch-072.pt","sha256":"a2b07bea8609ae0940fac4e88fc6549e9b2addf110f07149ba265496859ad81c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-000/epoch-072.pt","sha256":"d40a24233a9e7f20ee7ddb6dc318f37fa381c0c62fb89f84a9e3f88498775837","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-001/epoch-072.pt","sha256":"c12ca6e64ee9f4643e4e1cf1480a4452a955ddbbed4a50da588f9009b4a939bf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-002/epoch-072.pt","sha256":"3bc347cc3a7dbbfb114630fc0e5ba32e02726eab4b7e2af7b979c145d6d586b5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-003/epoch-072.pt","sha256":"cc7d5eebd4f09c90363f4ac16a0739dd8a923c8be1f394a1681e2ec0c3060cca","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-004/epoch-072.pt","sha256":"2dca0322b0f7860e36b37e882a1f665fd4bbd63fe3bc8187f74d1c4b7b185b95","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-005/epoch-072.pt","sha256":"2836bdabe814c297172b733f2c7d081b32ff132dd5e6d84d372784b7efbef1dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-006/epoch-072.pt","sha256":"2bce0298bd13de03d9be4a1f3561c11dee19f907c1483d94590c8ca2ad03be03","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/node-007/epoch-072.pt","sha256":"b1de50eb3a1a7235058497d66ee746b11c9069d96dad3cbf47d4c0e0dd8a8a75","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001/continuation.pt","sha256":"84a15d7a72e305356e6794e50ce8ce64f7ffc0dbf506fdb921d19c9d9729fec4","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fddd30b411d0c9f1902a7d91d8d0e1eeecc9d195be7506545203d87876e9ea76","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1fa2f89a26cbbed64cb297df6248e4ca7da3b1e0652b25e6b43c79c513a591ac","loader.blocks":1593984}}},"awc8-wsd-h80--0.004--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.004-s44","horizon":80,"attempt":1,"job_id":"2482123","recipe_identity":"f219c554de6146cebcbd107939d10041be34e0ba920866d2745143913d44cb1a","loss":3.4009676760500986,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/result.json","sha256":"ee85d21f097e6481e0a1cc8a7479f5aff12fc4c839f25019f81462493a983fd6","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.pt","sha256":"268c684d2fdc0084ee929c60d0e4b049e1a25b8908bee71e1cdf60e043385f85","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/final.safetensors","sha256":"cd23acf5c6bec9807966797100bddd8f2c84ee17f5247cb514de0544b93d6dcb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/final.safetensors","sha256":"2651e63fd12164b7e25988cacc540fb2a9fe085ff7901db72e147ed12279c94a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/final.safetensors","sha256":"364d22d38a08272dbbb81f3cb0f7468b0c58babf3ed0225f8aa3feef1a8094dc","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/final.safetensors","sha256":"993f518a75aff8d719f1b3c884b05e1ced6582c1f602eb04485a3290052db0cc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/final.safetensors","sha256":"3ce4377121d67afaa5eb3280132c5aa172b0c4a89ff84f76d2c66f569d53617e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/final.safetensors","sha256":"2651e63fd12164b7e25988cacc540fb2a9fe085ff7901db72e147ed12279c94a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/final.safetensors","sha256":"364d22d38a08272dbbb81f3cb0f7468b0c58babf3ed0225f8aa3feef1a8094dc","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/final.safetensors","sha256":"993f518a75aff8d719f1b3c884b05e1ced6582c1f602eb04485a3290052db0cc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/final.safetensors","sha256":"3ce4377121d67afaa5eb3280132c5aa172b0c4a89ff84f76d2c66f569d53617e","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-143.pt","sha256":"75bf9fb30ec7fac408fc2e1b078cc5c01eff37466a00dcdcb933b448f4a997cf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-143.pt","sha256":"a50eb83d8865301ea9053c737bee209849edf64dd116c15c5362201b535b9576","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-143.pt","sha256":"24e04c9d4be8b0ae944320b425258eaf28d77ab7709fd2f74358e44b181fbfaf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-143.pt","sha256":"9f12e21ce1fac67ae3fdd0ff7f9df66509b1425ad105190a9529dfe55d39ec85","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-143.pt","sha256":"f8d1dc72a183ca30d2c292681056448c297b60832fbbea0eab68a5391880bf30","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-143.pt","sha256":"176cad0aac7c74895d0cb2824a7325faaeccf82e3124939c3ffa087d64957fa1","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-143.pt","sha256":"83403c1de882a2b07095e9ba37dc6692b628d82850a7acc467aa6c39e897b1a7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-143.pt","sha256":"b26d03b063c6f2369fb2e79f0932c5494ad5e495fc06077c79472ff0a1e77b3a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-143.pt","sha256":"c1f0fea8bd17eb32f1072d83f4f5ad7ecae40c1903635aadc5ea10bcec2428c7","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"edd734d87140941b2e2520ab021751adad26203f4dabb84a2d82cf8bdf6ad48d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"c7e412dcae5342e6c12d41072d443d45847c4552ad4fc8ba870f6cb8ce66ba2c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"d7a36144eaeba630f0f3eb657be3e03022e5da2224c4fa3a3aa0c87c73db270b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"5c335724f9af4a0b51fe19e8d5df24ddd3a7755b8fad022142edbbbdd5d1ae7e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"1e5ba30ad70697e5cd1c4626d3e753fa26ede5f6aab300ab65ea01cfaff3804d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"98a67a787fcc7e527a2e369c7827cce51d6cc9225f040feb6fd89e0919254bd3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"e1152bc8c409c7bfeb8b224eed60fc0ff98e8fffe8e7cc4fb19c0750ae04af22","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"23691ce948d68b38e2e6f5214d9c5e22a844e9752606c53311f0b359558bdd67","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"50347736fb52a96552f745692f27511267b8e651043b75ae1ee5cd33780dd4f6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/environment-resume.json","sha256":"5eca38b4dc8a5b6f3936a113847aa7dec45d8f79bca31d4307eecdc4e2e07315","bytes":5344}},"requestSha256":"88aefeeca01fa3531403c86458a1953b78a78e3423d5732eae0e35d7a9b56eae","metricsSha256":"c710b7b5ca12a5f921808b83eb605e5321adcd06e31411efa904fd04d1cf639a","continuationSha256":"44b3b1714493bbae0f02db4eb8dd4d3dee5712e58532ff8819d3de4a8f78cddd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/epoch-072.pt","sha256":"edd734d87140941b2e2520ab021751adad26203f4dabb84a2d82cf8bdf6ad48d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-000/epoch-072.pt","sha256":"c7e412dcae5342e6c12d41072d443d45847c4552ad4fc8ba870f6cb8ce66ba2c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-001/epoch-072.pt","sha256":"d7a36144eaeba630f0f3eb657be3e03022e5da2224c4fa3a3aa0c87c73db270b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-002/epoch-072.pt","sha256":"5c335724f9af4a0b51fe19e8d5df24ddd3a7755b8fad022142edbbbdd5d1ae7e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-003/epoch-072.pt","sha256":"1e5ba30ad70697e5cd1c4626d3e753fa26ede5f6aab300ab65ea01cfaff3804d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-004/epoch-072.pt","sha256":"98a67a787fcc7e527a2e369c7827cce51d6cc9225f040feb6fd89e0919254bd3","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-005/epoch-072.pt","sha256":"e1152bc8c409c7bfeb8b224eed60fc0ff98e8fffe8e7cc4fb19c0750ae04af22","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-006/epoch-072.pt","sha256":"23691ce948d68b38e2e6f5214d9c5e22a844e9752606c53311f0b359558bdd67","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/node-007/epoch-072.pt","sha256":"50347736fb52a96552f745692f27511267b8e651043b75ae1ee5cd33780dd4f6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001/continuation.pt","sha256":"9e9bd35a89726917ad03b00c09bd09cbf1464aad164e0d615ddc61318d9f875d","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5017ad899ec20cdea7253aed987bf7acd3424b65a75506480d52571ca4ac221a","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.004,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.004-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f219c554de6146cebcbd107939d10041be34e0ba920866d2745143913d44cb1a","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":80,"attempt":1,"job_id":"2482124","recipe_identity":"76f519163ac233eb6964fbf33c7e709daea62607161e95d3ee4d204b68e661c0","loss":3.3901543558150875,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"64f468993bf53c6c7f98cb977025dd8e2af3e333f75042b5a400709a8311620c","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"b05940461ca1136bd6c3d17d59ba05cdc7d92da3a5c16eb9f31b4448648ab465","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"a2fe9bd5155537fc5850053e5f8707802b571728500598b0972dc99fdb6b3f3f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"c26ee7b625dc926e03cfcf0580e8adf681f7cd0d257b511874daa7d91bc419ab","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"9eb7dae52d9342adf8ab6623c11575e72861d48648d3ac17251eb43974f811ca","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"5b2f264fe8dce3461abc84ffea3308aede62563ecb273d56abe761e861b89641","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"9593e869d906c597818c72d2b0887c93fab50e2d835931fdea7ff3be5245bf8a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"c26ee7b625dc926e03cfcf0580e8adf681f7cd0d257b511874daa7d91bc419ab","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"9eb7dae52d9342adf8ab6623c11575e72861d48648d3ac17251eb43974f811ca","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"5b2f264fe8dce3461abc84ffea3308aede62563ecb273d56abe761e861b89641","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"9593e869d906c597818c72d2b0887c93fab50e2d835931fdea7ff3be5245bf8a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"2a9c90ef3e897ddfbf36c2b62efb7292dacc278275f37d790fb756512be08c9e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"2532825d2401213902d6d017a40ed86eca84feec4bb7ffafabffb4a0c58e9727","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"755125eaa93f1305398ec4389ee158a6d7ae4acde2c3438869e8e4a057c6976c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"8cb88ca384c8a14c13a3945178c933be767b1940e5d3b61ac40daa93c9818cfa","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"3374e3c876672f7a542c7606301142aba79ebabaeec9b0e1d4448b224b673dcb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"8556484f6a3df53aa0b82cd812897a2ef90e4f2128f82f86f552ccbd5df9e9c2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"90d24a0085da404b3fa193f64402c16931401a5050bfe9bde54fd69210bb432a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"8aa9eea638ab1eed60e565c9b8305d4917e9f74e3accb9f134639a3ea6a19776","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"490c3543eaf1c30944b8838f5efcd242b261d52655c8ce79b301364144b0d3ca","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"fe20269a017cf7e9f87d29133253f946a20caa8c69702e8bc2a4c8442c57f483","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"f0462d1061975650b8add703ada21a8fdb8c0b4c6d01bf67acb38a040aba0c08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"513b8ba32b38ee07c4956329524a86c1d0b2578b5dc644d19d898fb1e5047934","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"b6fde3581cb43cf0166ff9ddbfc18a58c24a2a8d1ca8aa58354159cc88cf0757","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"7d41bdc1688359ff39a5656794b36bd7ae46af856c527ce811989a7b3fe7ee07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"e8a0c8df455031261b40f8e3c344d4fda5fa865c3c72f7ae706bc4823fa082e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"2fa1ddbffbac321d585effbee3b750736892fa5f0195f8198bd32098101171c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"3bed072c5ad8337986d8a75cc2229c2a80dcf98d2625e7e270488ee63d8d22a8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"a5520f79c639c87ddf2cd83a9a5c0af5b939f4be06bd27d18a6d6939a547cf40","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"1cd89c073d8bdca886d3f01a94e2c7af82675476a412ad94fcddcf49601c5e91","bytes":5343}},"requestSha256":"e8b9a6621e12c7c0dc46f7d59445ca696aeddae4948bacc2127032d8c4d9b080","metricsSha256":"bd60f97286b6182af931520aa75e64df15c0db632bbe589a84ee979816ac1f93","continuationSha256":"23ea6897fc04fb5c51a5552c7948951b9af99cd2503721fc6a3a2a37ad217f8c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-072.pt","sha256":"fe20269a017cf7e9f87d29133253f946a20caa8c69702e8bc2a4c8442c57f483","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-072.pt","sha256":"f0462d1061975650b8add703ada21a8fdb8c0b4c6d01bf67acb38a040aba0c08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-072.pt","sha256":"513b8ba32b38ee07c4956329524a86c1d0b2578b5dc644d19d898fb1e5047934","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-072.pt","sha256":"b6fde3581cb43cf0166ff9ddbfc18a58c24a2a8d1ca8aa58354159cc88cf0757","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-072.pt","sha256":"7d41bdc1688359ff39a5656794b36bd7ae46af856c527ce811989a7b3fe7ee07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-072.pt","sha256":"e8a0c8df455031261b40f8e3c344d4fda5fa865c3c72f7ae706bc4823fa082e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-072.pt","sha256":"2fa1ddbffbac321d585effbee3b750736892fa5f0195f8198bd32098101171c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-072.pt","sha256":"3bed072c5ad8337986d8a75cc2229c2a80dcf98d2625e7e270488ee63d8d22a8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-072.pt","sha256":"a5520f79c639c87ddf2cd83a9a5c0af5b939f4be06bd27d18a6d6939a547cf40","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"57cb17aaed10b6f356e44c9baa814977b95a5632ca0007e62153829d34422329","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ab277b0ae79c238e590aaee645ff1ed1a6de612cad610d4c4cced23aef522150","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"76f519163ac233eb6964fbf33c7e709daea62607161e95d3ee4d204b68e661c0","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":80,"attempt":1,"job_id":"2482125","recipe_identity":"b38cafabd16cf8321a81eb37d669a14db06d964aa68e49834a7797228514ba44","loss":3.3991775153865236,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"0bda8c179798a3aa622fb1e449c2ab0613632af514f2a472e7cddc5d73877f8b","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"84a3b0ec03f5d784f0e333ed00c211a52ba1100c77c323aa63a8436d025eff2e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"77c9cd2ffb9917fa8b5b885a9a00ce617df6732cb080838ee3b41aaaa9a6e7ad","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"e9faa3561e553990825cf365efbf4ffa22ee4e20fc07849589404404478763e1","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"824c02810a08d36779ca8572e15ccedcb58579483f560effbc4ec814f894d849","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"ced4826f6d11cbf72ccff3c1ce56865691665d3f795a0810cb08834c46458f92","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"aa14afeb8797f2fdda3e4c773a7bca5ab4a0db2ee0e666092e0c32f9c7e15022","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"e9faa3561e553990825cf365efbf4ffa22ee4e20fc07849589404404478763e1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"824c02810a08d36779ca8572e15ccedcb58579483f560effbc4ec814f894d849","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"ced4826f6d11cbf72ccff3c1ce56865691665d3f795a0810cb08834c46458f92","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"aa14afeb8797f2fdda3e4c773a7bca5ab4a0db2ee0e666092e0c32f9c7e15022","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1492e8591ea2fc100602d98a5637cbfb0fa3d19408ae6e95c58429df3246b11c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e3f77aee6c53fcc38236fed290efb61fe3d0ddbcc1503bef44ee474ab16a6791","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"85e98cce57a4dd8f11562b462498f5a0c6cc0fcf4444b6ae84188483422598a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"f404a5301622ecc696d1452264c5ff6af87265b3fc2ee88f31a1e028d33329e5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"4bc2700bff93acf1b45353d7e77c48bb1148dde20f73db5588afcefeea260c49","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"fb0d7c27ae5119616826a4185de79cff968d787632657b62bdd0a3b89d853587","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"e7601f1ba9fd632900bdb1130274d4b1e06e2fd1e6a936a9bcb156996e17e82b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"ae449f74fd50272b8ad10a2ee60521d06f3abc5c9ac96d82cf58288453ed248a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"f6da5a05b49d5a521873653ad3216029a12d900637a0b345fb08b3fe7636a4f8","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"4d3d69ffe618f7f4d525bb10169cb0fa7d07c77f3e5cbe5e2efb0b2aef049171","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"fa4a2eea1e4b3f8542e77ff42af08d2f8edd26cb2bebf278d3d703e6c7bcfe08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"04bee43a1e3d732b9ebf46c85e3dc2ad79da192daef539feaecd77bb6f37d1d9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"69748e30db69badf9226303a6da3207869c10a555e691f19de030a8433756404","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"3fe530e559156c6f78ceade1b6d8ae07cbdf9b489431cc9d8f67f4e5a7af819e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"6a6456eff7d372823708cf2497bdde29dc2943c487a814cc6ddcf6d9679f82b8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"7cc6c5c5dd952ea3637b8609dbdc30aa407d73c708ce6260fa5e98b97fb036e1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"c41f8136b858fa0a7dffe9f2e5aae08ba14f95f7fda8c26f202c347fa5f241fe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"b52320d6148e0255efb57dd72b320866919e411fc5558114e601c59e0c32cdaf","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"1503cc40f4e72a8e237328a6416886736ddad079784eb6ea6ac7e3967c602685","bytes":5256}},"requestSha256":"03a95b60cc480e673f5fc09840999bd6cdb284321817759755b29bcf01ad7bb2","metricsSha256":"6ee1cdc6307abfdc7ad432077e17bd0624c50241ee408e8d0c65d6f7cc0d13cf","continuationSha256":"f09ea63549c5a2bfa83462531197de25ee235bebc59a43db3c5213083339c5ca","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-072.pt","sha256":"4d3d69ffe618f7f4d525bb10169cb0fa7d07c77f3e5cbe5e2efb0b2aef049171","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-072.pt","sha256":"fa4a2eea1e4b3f8542e77ff42af08d2f8edd26cb2bebf278d3d703e6c7bcfe08","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-072.pt","sha256":"04bee43a1e3d732b9ebf46c85e3dc2ad79da192daef539feaecd77bb6f37d1d9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-072.pt","sha256":"69748e30db69badf9226303a6da3207869c10a555e691f19de030a8433756404","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-072.pt","sha256":"3fe530e559156c6f78ceade1b6d8ae07cbdf9b489431cc9d8f67f4e5a7af819e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-072.pt","sha256":"6a6456eff7d372823708cf2497bdde29dc2943c487a814cc6ddcf6d9679f82b8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-072.pt","sha256":"7cc6c5c5dd952ea3637b8609dbdc30aa407d73c708ce6260fa5e98b97fb036e1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-072.pt","sha256":"c41f8136b858fa0a7dffe9f2e5aae08ba14f95f7fda8c26f202c347fa5f241fe","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-072.pt","sha256":"b52320d6148e0255efb57dd72b320866919e411fc5558114e601c59e0c32cdaf","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"e453de48fec791e58cbd813412eb1028518e7e3b613bd08cbd592c0a8fb95785","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"916802bdb54c05fdf34bce89371003ba29b1ab7dbbe16e56946184db4bd70b39","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b38cafabd16cf8321a81eb37d669a14db06d964aa68e49834a7797228514ba44","loader.blocks":1593984}}},"awc8-wsd-h80--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":80,"attempt":1,"job_id":"2482126","recipe_identity":"03cab47858b55e02e969f32ba79b9b866e2a4ffab84e8ee29e5784a82f3df72a","loss":3.3966294039874243,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"8abd1f9a9efd05be3665d86110fc951d2411be0dee3f19065f44fd0830b0f513","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"057534f20d034075d35347328e770eda90ab90db55fc2dcab5819bfde937fb2b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"afa55bcdbad29443e082c953a080ac0831652acdf7b5283af21b67ddd4d8617b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"614961603db51f6e8282fdf3bf5418d5dadda8a3d51fac029407413f775f3c58","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"3ad7b26df766e7b6439e1aa2b796a58d797484d4889d07a6081ea4ee91378c79","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"4d01277ca9e69df9a893a525024fd73611cc2c54e1421af62f140148abab652b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"bdd3505771d74f30f55d3ed13b2e95f12f40cff19f7a1766d8dac1330ac6e455","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"614961603db51f6e8282fdf3bf5418d5dadda8a3d51fac029407413f775f3c58","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"3ad7b26df766e7b6439e1aa2b796a58d797484d4889d07a6081ea4ee91378c79","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"4d01277ca9e69df9a893a525024fd73611cc2c54e1421af62f140148abab652b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"bdd3505771d74f30f55d3ed13b2e95f12f40cff19f7a1766d8dac1330ac6e455","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"c36ed0f07aef24b23032e56ad073d153c146a81c073f8343b588ff02e87a568d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"661ae6123a5e831679d006343a8246086d10810d957db086ee4ffc573f1cc2df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"4134eda656693b6bd159359197a7b99a58864123164a0ba6aae4ced6ee083663","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"96fa2d82fe6684448f630081103b9960219b0b560d25831a5ba58e423b936367","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"0358e9f67881e0f9accebb8f156117b1aba32f1b5a7a5852c7e0fb5479282bb8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"e9bbc9cc2546c723092908a2584366e9e1f1fb504f35479d1f83f386cad064ec","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"7043ab42e02d101bfdb137c9bb952de79ee579a22d15dccdd8fa24f60f7ccb1d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"5fcb7b2ec11fae66ad4f9e39036eba349a6854b9b0a4bee91c04e459340e0a0a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"2f3693a357b5669ee2b39263ef6ae56d034ed5a76bf8672d0f7706919921bbe8","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"d8e8c4bd87e290188245252a082b38377977b7008f6aafcc08285191d0469169","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"ba269c7d1a3e6fd6eebce28ae61d5be8994cb4d4459893a973abdf992d7223e1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"443b3ff78606bbb16da279ff2da0131ce0a729c5afcc7d9c242c15e7ea25fad9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"355c486cd54dbcdcdc6586072df4b3926de6c7396747e28d49a44862ffcd03d9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"0456208e08aa1261bcb43ca3ad14ec0ff45f45aa244b74f8e9a2546604c27f01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"6dc3dc8a1e0351dcff3947e88328862733d47380bbe51254e591d947593da2f6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"49e6621b4600f0f59ac11fb6f610a5b793f7966d50951a8f7ab14c805caa9a25","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"ce280d34f6074ca401024ca0b67810ec316dd8e9848ec08de94ef87a3149b22e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"032aff9aae451f6634d6fc9cefb593507dea1ad07cc0cddbc8a1d1bbaf91dedd","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"c4e104ea832ad249d72d2a6f72505ab95bdd6a60af5cffd7c046591a5b6df559","bytes":5315}},"requestSha256":"7d0d87f67f8e6a0c9711abd23b08528436ff57322ed4d19301abfa0d34178e8f","metricsSha256":"0bb15cd71d4d826a02b4208ff6b2e3404107cc24cc9a4833f1ab4962a03cf376","continuationSha256":"c97f61ef3f88314d4eba536f768413b8cc84ce9ab47d75ef209e67c4058eb94f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-072.pt","sha256":"d8e8c4bd87e290188245252a082b38377977b7008f6aafcc08285191d0469169","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-072.pt","sha256":"ba269c7d1a3e6fd6eebce28ae61d5be8994cb4d4459893a973abdf992d7223e1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-072.pt","sha256":"443b3ff78606bbb16da279ff2da0131ce0a729c5afcc7d9c242c15e7ea25fad9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-072.pt","sha256":"355c486cd54dbcdcdc6586072df4b3926de6c7396747e28d49a44862ffcd03d9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-072.pt","sha256":"0456208e08aa1261bcb43ca3ad14ec0ff45f45aa244b74f8e9a2546604c27f01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-072.pt","sha256":"6dc3dc8a1e0351dcff3947e88328862733d47380bbe51254e591d947593da2f6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-072.pt","sha256":"49e6621b4600f0f59ac11fb6f610a5b793f7966d50951a8f7ab14c805caa9a25","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-072.pt","sha256":"ce280d34f6074ca401024ca0b67810ec316dd8e9848ec08de94ef87a3149b22e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-072.pt","sha256":"032aff9aae451f6634d6fc9cefb593507dea1ad07cc0cddbc8a1d1bbaf91dedd","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"1247c575454fa6075832829bc680da5c25ac134418f9b211e2ce31722ddfa4dc","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c844d0a29cea7542c33f01c70c5eaa88d94f47b11345237b6579635b556d590c","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"03cab47858b55e02e969f32ba79b9b866e2a4ffab84e8ee29e5784a82f3df72a","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":80,"attempt":1,"job_id":"2482127","recipe_identity":"703838c8661c7739dda29096ec11c1ab9ac0b9463c57398b830e6a5ef54531af","loss":3.4125951797725493,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"6c6314dd2d4edc3c4bb633390fadcd74bbb11847fde9e6664d4b4f043d6384bf","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"935af85ec6964808c9b6777eb7dba024007871586eb4fc616fbf8b7e3f392278","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"b87036258286b711171e63cb430ddb59b48657268ac81a767ce7de587d58c580","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"2ede870b9b7c580730f503671b09bf6a1d9beadbb5cc3f498e39bdee8763c605","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"2152fb4c0887d2bde0235cb9a21cfe653692330e4bc3f04443250526627dd96b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"85bdf41c416b796d68b40911b360c0fafb1d96aad53e03a7ccd6f64f22c5a8bb","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"6e2041bd78418d546388736fd6a75965a47da476875311783e81a9769b8ecae5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"2ede870b9b7c580730f503671b09bf6a1d9beadbb5cc3f498e39bdee8763c605","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"2152fb4c0887d2bde0235cb9a21cfe653692330e4bc3f04443250526627dd96b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"85bdf41c416b796d68b40911b360c0fafb1d96aad53e03a7ccd6f64f22c5a8bb","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"6e2041bd78418d546388736fd6a75965a47da476875311783e81a9769b8ecae5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c348217d0d02845b469c21993bf745e2693bc49b78ac1fc06b91223264c2fbf3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"c91e2fd958936618746c692d71fd2333c52ffcbdff1c09b071e18129c06e571e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"91d37f3cd4483de0cb8f272f42a57d0df5b45e4e4763866c5b760099447a70ff","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"6df6c499694295de032d4bc2387985e9f0887f6cc0671a52f661269827d0b5ce","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"bc1fb668d17594c926e9458c38beaad3d41023e7eb4d79bcb42ae94fe5ddd62b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"e80bb44d911c3ee031334dccc351b6479df2cccd54e8040618ef721b81a32e6c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"36f7eddb8375d408be05aea6d205431aabd0bcac02e3eeefb2467b09cbafc286","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"2f115b5be10ac40fde3be6094e234833f3ea4c3815431cb17f74c79e4c7676bd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"56808e711519a1205bf19c320045d8c38fbab238bb86a5bc4f6ff251545d97ea","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"3514c7854042c19ffec15d8e4b6b2683a9b2f633b06123f3486b7e66ec6ad5e9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"dbcf3cd99798eeebb7018961139fcc4219ca29617afe26847aa2b9ef4947ccd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"33d3430afe67735f94d28a7f228ed10eb4b18b6c94042929843e01a448a8e9cd","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"750bcc37fea1b026d9d057ebcd22fc6e1467fc742201e14f47c3b001fc09d5d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"238c711be143491b4e3b6027e630ff1797d5619e467d8ff03868c04d9ec7016d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"c09dbfdc7f788c71733173c22c47db194a7f784aa6e15e6b918d8b23c7a79a9d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"945f5714f43c76d3ff4019f0fac54421586bc0182419473bce513b4346907ee2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"34a09e81f768263acc603fa64dd5f5cf2fbcb68fdb43b425caaababcb9f40878","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"990ebba1094b5147a50cf95b3d804cbd4bc77c36ea1306237c9f585e70935f32","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"4984ed1972ba51babdceef2e3e7b943477e7c841653612a2d0e571e9f859aee3","bytes":5344}},"requestSha256":"2fddde656036f5fa3ad50408b713ae1188ab87e5dab4ddd9334a5d06b7166787","metricsSha256":"d029cfd3664242295086e5a5e5f6bdc0e28c3aaecb71118f4eb5c15c13d59f5d","continuationSha256":"3bbf1736ffa6c68b4595fccf92dbde0cf22147a9ed9e6cf4198c5459945adfcb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-072.pt","sha256":"3514c7854042c19ffec15d8e4b6b2683a9b2f633b06123f3486b7e66ec6ad5e9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-072.pt","sha256":"dbcf3cd99798eeebb7018961139fcc4219ca29617afe26847aa2b9ef4947ccd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-072.pt","sha256":"33d3430afe67735f94d28a7f228ed10eb4b18b6c94042929843e01a448a8e9cd","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-072.pt","sha256":"750bcc37fea1b026d9d057ebcd22fc6e1467fc742201e14f47c3b001fc09d5d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-072.pt","sha256":"238c711be143491b4e3b6027e630ff1797d5619e467d8ff03868c04d9ec7016d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-072.pt","sha256":"c09dbfdc7f788c71733173c22c47db194a7f784aa6e15e6b918d8b23c7a79a9d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-072.pt","sha256":"945f5714f43c76d3ff4019f0fac54421586bc0182419473bce513b4346907ee2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-072.pt","sha256":"34a09e81f768263acc603fa64dd5f5cf2fbcb68fdb43b425caaababcb9f40878","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-072.pt","sha256":"990ebba1094b5147a50cf95b3d804cbd4bc77c36ea1306237c9f585e70935f32","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"b368f7a3434d1eec3d24d4a5134df7c98823e9e88dd82d1526a17d60ec8e612b","bytes":1958918101},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"095acec6c28d4db570adbf8aafc6d3ec619a58c775cae51c5d7c865de6cd51bd","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"703838c8661c7739dda29096ec11c1ab9ac0b9463c57398b830e6a5ef54531af","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":80,"attempt":1,"job_id":"2482128","recipe_identity":"97ebdce4b9b90a0b5a45a1dfb2ea127a43c8b64f1e01c41c53aa46551053811f","loss":3.4147759569561096,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"dd263917260a5ae8f62e8720f7e2047ffddce22e3e8c6c877d74fac50297a8df","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"4ca9d5493ffb55be5e74a8755afa07677b93570e259abdd7c47834489f37bde1","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"06fd5238eb69bd55c866a672551e0c5edbb7e0126b4ff8ddd3d2867ccc1801fc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"84d5937e4bb75db3d6064a04f020b23542accfc2daf92da640cc1e8bd11e8c08","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"f70e8fb4a64d477bc6e06395fe4ac4c9a91480bdccbdbbff979c64165eec2d8c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"528034c0874054dfb61351c8776f8279b3fd147f031057c1634086bd1cbf0ff0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"9a951cee56a7e67f504ac954ba55b64e8ad3bbc56b8e4a5a7d93df75701319d5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"84d5937e4bb75db3d6064a04f020b23542accfc2daf92da640cc1e8bd11e8c08","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"f70e8fb4a64d477bc6e06395fe4ac4c9a91480bdccbdbbff979c64165eec2d8c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"528034c0874054dfb61351c8776f8279b3fd147f031057c1634086bd1cbf0ff0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"9a951cee56a7e67f504ac954ba55b64e8ad3bbc56b8e4a5a7d93df75701319d5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"678f11fb48e68e240a7e07c24185857ed25de4d345dcaf83b6bfe02925c144e5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"e08dd502f93353f7a9350d80acf8d90a7b37027d505fae0b280cb86f01fb49e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"2e5908220c432eb6ae964fcd71577bfcecdd3b04893786795823118290a0436e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"c686124b10d6fcb2a8fa56ec74aaf669e279819c4ef5692c528c27f445463248","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"8370d408864673eab0e9e9d8e2e8c0f5646eec2a9f82e96f562bbb40c597c77f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"7259b058f58444697cb6d61fc2a543e560d47d0fd46bcf470dfae4f6742e8e2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"d9107fbcf460f0d93770f10f360ae3f54dae96116621db3267e6d32662cd38a2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"3e2ca745e0555142c0a6be786713b9e22118d0c37d541a186667669fd0c3f4f1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"acc9d33d01e0cd2e839c211082cbeca972a8c6fe528170631ca6b1d4d1ad2dce","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"8d82d8df2c16a4aad045d3ab6b1ed8afd2064fef7635279922930855abe1fa7a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"cf82cc0427952d9cecf9abdcc697185ff56ad0f3d74b478f8f9e5c015dab45af","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"3a440ff76cf55450df1371dc3facfd2dc1c44041a2f0ba590c58f461ff82c0ee","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"13392f95a92d6f5d9978159c4a8ab73ebdce24790011dacc2385a7d9ddc6f4b1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"ab6eb43014b2d33a5e224f8f6cb62e2598a970f87d11ae5d1acdddd5d3a2ef16","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"fcb13d2a20f351bf409bd6c813f6152842c1a83343a640171c4cedfa62b15d92","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"9cdf99e417f3a243ac28e7f3f7d023f064b4cfec5950c181e562e28b0cbbb8ad","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"9cffa9f4391746cab1187217d451e9fcb22bb07d352bfa712569dd5e9d056451","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"2d3329d8ea65b9809b9e19d17ae3d17c370805ae74ecd4e6027d63b41effdb55","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"f7be5e4c6667b7b71c975a1bf9614689ac0861ff8e7e738c85f18a07c044ba07","bytes":5343}},"requestSha256":"42d91c5b59743bee6a8c41418953f4d5ed16171408c0d69715a1aae4e63e2e64","metricsSha256":"210e8973debf89654d1c95618239693f77c21a52277252aa2ff317bec379b08a","continuationSha256":"12259255faed3f2b3579d22e531e32c282938cab6e92ac69f435d32ce09e90af","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-072.pt","sha256":"8d82d8df2c16a4aad045d3ab6b1ed8afd2064fef7635279922930855abe1fa7a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-072.pt","sha256":"cf82cc0427952d9cecf9abdcc697185ff56ad0f3d74b478f8f9e5c015dab45af","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-072.pt","sha256":"3a440ff76cf55450df1371dc3facfd2dc1c44041a2f0ba590c58f461ff82c0ee","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-072.pt","sha256":"13392f95a92d6f5d9978159c4a8ab73ebdce24790011dacc2385a7d9ddc6f4b1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-072.pt","sha256":"ab6eb43014b2d33a5e224f8f6cb62e2598a970f87d11ae5d1acdddd5d3a2ef16","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-072.pt","sha256":"fcb13d2a20f351bf409bd6c813f6152842c1a83343a640171c4cedfa62b15d92","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-072.pt","sha256":"9cdf99e417f3a243ac28e7f3f7d023f064b4cfec5950c181e562e28b0cbbb8ad","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-072.pt","sha256":"9cffa9f4391746cab1187217d451e9fcb22bb07d352bfa712569dd5e9d056451","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-072.pt","sha256":"2d3329d8ea65b9809b9e19d17ae3d17c370805ae74ecd4e6027d63b41effdb55","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"2458efae52b17fc47d370a38504a3834e757b1cdc76b612f660390a90f941a19","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e9ee41b7330175910f22cf2570f2f3bff8208bd387bf18336e9456431a44c54e","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"97ebdce4b9b90a0b5a45a1dfb2ea127a43c8b64f1e01c41c53aa46551053811f","loader.blocks":1593984}}},"awc8-wsd-h80--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":80,"attempt":1,"job_id":"2482129","recipe_identity":"590a0070fd123dcc19744fb2ad01a50aa8cca1f92488804ece0544bc50bdfbe5","loss":3.41218367876118,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"f0609752b974ebe302f956801254f5ca058e24306238ccabc3ba8cfdae3d70de","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"3301b0fecd1608f9a43ed5e65affbf9b03fda1989161c6d7670c7d3a035d53a4","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"a30dcad538e94fc78cec1ebf8c8435f3c4098671cd49ded13fc748cb9ad7e352","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"969bfdf5bfc9ea373d7c09af32195a1e820cae9cf8cba34c5f0f878453776881","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"84c6923aa8486d1bef0c47d01aab126255530a05d84c6bd716b2f91355ae18bd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"651a893b38aeea86e82c5e19b77a08e70b528ede638c7d5616b2373c32bca897","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"9109e12a6b5e475ed57516de2b6e6c1969cbe95349e0a5a30f69008d59bd6148","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"969bfdf5bfc9ea373d7c09af32195a1e820cae9cf8cba34c5f0f878453776881","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"84c6923aa8486d1bef0c47d01aab126255530a05d84c6bd716b2f91355ae18bd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"651a893b38aeea86e82c5e19b77a08e70b528ede638c7d5616b2373c32bca897","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"9109e12a6b5e475ed57516de2b6e6c1969cbe95349e0a5a30f69008d59bd6148","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"f0529862cfaff35a04380f7e72b9a5db6c4680ae84088ed04c28ec4ca38cba4a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"b2eca46b0b2c10038dc88ab82f417bde9ae208086fca8a039d0f1816f343b77b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"0fdc44c6ca8c9ef9c5920ca68935736cd926bd0eaecf5efe1821a333ecdea0f4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"d91b4c22ba58fa8147bc4c48943d85a1c8243d5e1a23a99eb591c278d3d0690d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"0c2b9449e94bd895ad2b0045b9da1c20157a54312d903dfa54d0863cd459564b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"4567eba83df13d7b25283d45c1d5822140ef2712e7a7215caf817b1467982bc9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"61fb4ab91ab05e77cd9da2d8b899f7a048c18bfb9a99161448f4103c2cf74228","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"6a87785700f929b8e27e9ad8f887bed86620b22e3cce6fad37941e84ccacb372","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"d27012f653e2abbbb14615772a74a7ebeeeab9dfe603d7cda292b1b4e23fde80","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"1eed95c6f0950087a330412d35d6e6cbc68b118e439357d406e01ea4e515760e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"e4bccb31b73ec04499e016eef5770ede0386542add482d6350fdf4a5d495ca48","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"284261ba99aa2506c3f20754dc58b27c4f07ab0e845d4bfd3cd17554e569677a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"e06b78ca36b2e054a01550df7b0a30e072255b58bc8930b79b79fe61a99ffb49","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"14327ec46293fd6b0233201b2384f720b7d68bb463c4bf859316e6576740dd69","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"30d82771a55cc2bf03ca7d6df97a21a38e53c015a2aafd0457fe326f38828436","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"2e45cee83ba937594b899c174a35ea1d8d9c93c3104b8fefa6f5eaf523ca4d07","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"f1f2ec21b4936e6148e4f1956e6e9a1e1b1db0b4515d9a57902c60896e269eec","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"8ee271ec3be83c154f9d3ae050bebd410b4b68a0c71b744c86c87d0a5dc352ff","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"0c0359d8dddf69f19bf97417c05a14ca41baad2f2f87d252a8bdb07cb91b5bbe","bytes":5258}},"requestSha256":"3ca998a45635a81077462bc07f0f7ecf2367f64b66031a2f21c6a3bc163c65a6","metricsSha256":"946d32cb5a0086a2d42424dd3d46ea475ae08b54eeb3031e255e4364af372b04","continuationSha256":"ace3c349313b971043674e59c924f8f6e922bdb4a78107efa399b2c43c5adb59","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-072.pt","sha256":"1eed95c6f0950087a330412d35d6e6cbc68b118e439357d406e01ea4e515760e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-072.pt","sha256":"e4bccb31b73ec04499e016eef5770ede0386542add482d6350fdf4a5d495ca48","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-072.pt","sha256":"284261ba99aa2506c3f20754dc58b27c4f07ab0e845d4bfd3cd17554e569677a","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-072.pt","sha256":"e06b78ca36b2e054a01550df7b0a30e072255b58bc8930b79b79fe61a99ffb49","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-072.pt","sha256":"14327ec46293fd6b0233201b2384f720b7d68bb463c4bf859316e6576740dd69","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-072.pt","sha256":"30d82771a55cc2bf03ca7d6df97a21a38e53c015a2aafd0457fe326f38828436","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-072.pt","sha256":"2e45cee83ba937594b899c174a35ea1d8d9c93c3104b8fefa6f5eaf523ca4d07","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-072.pt","sha256":"f1f2ec21b4936e6148e4f1956e6e9a1e1b1db0b4515d9a57902c60896e269eec","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-072.pt","sha256":"8ee271ec3be83c154f9d3ae050bebd410b4b68a0c71b744c86c87d0a5dc352ff","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"8063ef8ba7b4e43ea026e054bd4c6d159a1f54aa7d048b99309f9e5fe0b367c6","bytes":1958918037},"source_horizon":40.0,"target_horizon":80.0,"epoch":72,"step":5604,"cursor":717312,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":40.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[72],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-40/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a365bf3a61e7c80eab29f61e0e52cdf05644b0a835a131d7c459ea03e5ddfce5","loader.blocks":797056},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"590a0070fd123dcc19744fb2ad01a50aa8cca1f92488804ece0544bc50bdfbe5","loader.blocks":1593984}}},"awc8-wsd-h120--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2482589","recipe_identity":"e14d9486565665d2ea54939cb9db23cbf3e05b7c089029d6d31a61b2009a07fe","loss":3.5302565996856616,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"85ff98728866aaffc406c10b75e1aaefd8cf7d56d5ff471c412872a2d8ce440f","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"815a3700727dd390a1b27affb5d95974617ad94531088c8ca745c863a5843f57","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"bec720ebbfb46f6591fc07b8280b6cbe182b9a2d6cd965d487586ae9a727e96e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"6465bc3c4bca07ce40afc3a39bb58904401554af42ed8f6eac6fd4b57efd9597","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"94e04a4d09ecdeffd6763ae856b9ef632840164b48879414b34266367cfd3b8e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"4569ae86e6378bbe0f69ee5a37311b51e9adf2cb6326576d5f6ba960908c7389","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"e0982dc906f787f004117fb506eda87540af3d87863200024c26da1ac5def52d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"b7ccd979fc33c85a7ffea0923eecae90ed20b6a5f1544f86614d8ae2fdc45ff8","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"8d1bba87e936621d1b44461382fa07f25ecb5bfbfbdedb0d9d7bddc923c123d0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"39111fbbab1a5912ae7c931decc52bf07f51eb04c6660ca07ccb1a73829628ce","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"6260c36550a1dee90697184be494123f437556b00ac404c44de435859436cc01","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"79deebc99fbdb8d1100a5bdeda8be721689f9f8129e3c55200d91a3850e80680","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"1b0760e0c330272aa089d18b4ebd1455e5e7b5582936e5bc89a97f0f82a9d34a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"3063179051286258922332543de8b9f43435509dfa53cef73086441007b0e4c2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"1528b8314b24d76acad9ce9f332593315aef99eb691b002bf65966da23f6917f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"09a3d7a7b3726387a88ae5ba55850b66b63ad1e5881faf02188df706d9e6734a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"991953ca74fae39300db5d1f078130d134c868a9f6a2ad6c175c5fe760abdcd4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"dfa6f50851c4137b616ab8be2adae138e69fbe9f52b4bed82cdea61317f95c7b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"a04d0e6b0f99c87e86b304f117dab798a33f3a6ace5651d66cc8ca3edc2c6819","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"5e46943d2166070205be4969096bc9ea708ebe09268ddfb022279baab8358520","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"5ec8c86ad48c270bf98f0f956beed43a55b11dd7288e46e731705098688dc2af","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"204d67cffedeacddb3e350556b05677720ccb39fbd61469113089aebac104d71","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"eddaa33c81490b3e551bac6bea1c829a4dbff049dec885875294f01f81b42245","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"8f11daacf2f76b7fada67501929d5c3931f5241bf481acbc4da72b2a74bc80d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"4fa0b8842dc0f87bd6d063702c0673a464e6a5ee0380d6a38d5e0b8ce8c0981c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"bd1b5f22119e8e35beffd600e025791bdc8ef0b86840e306217475ac57d339f8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"3c6f684c409a892d7dcc64f17b16f80451bd162e651e95517e4d923586c1b5e5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"49a9ba0e7d395c870ac71db6cad403fc869d18a57e2d04a09130cdbae0ce536d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"e6d81fdf26879eb23c26377823aed1ed29abfa177753351fa130e00806c3a67e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"5014552914438b7a865187032d1d7d5e135fbdc0703a773aeb31f94b3271ce4c","bytes":5341}},"requestSha256":"4a1e4cafdf8064ffb62b36663dca4f6b40a84ad7441d0b9cbc07deda7da17d15","metricsSha256":"9bbda93bbb2c28948ad720f0674522a3aa945d11e0e8402ca7abe958d218593d","continuationSha256":"c62ad0c24ce4ed41c32c3189910dbc4c4a781d0c2c84381716763fd4a5e1bc67","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"5ec8c86ad48c270bf98f0f956beed43a55b11dd7288e46e731705098688dc2af","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"204d67cffedeacddb3e350556b05677720ccb39fbd61469113089aebac104d71","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"eddaa33c81490b3e551bac6bea1c829a4dbff049dec885875294f01f81b42245","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"8f11daacf2f76b7fada67501929d5c3931f5241bf481acbc4da72b2a74bc80d8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"4fa0b8842dc0f87bd6d063702c0673a464e6a5ee0380d6a38d5e0b8ce8c0981c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"bd1b5f22119e8e35beffd600e025791bdc8ef0b86840e306217475ac57d339f8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"3c6f684c409a892d7dcc64f17b16f80451bd162e651e95517e4d923586c1b5e5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"49a9ba0e7d395c870ac71db6cad403fc869d18a57e2d04a09130cdbae0ce536d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"e6d81fdf26879eb23c26377823aed1ed29abfa177753351fa130e00806c3a67e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"1a3ddbf1b114dec7be7c2e981fa35a079710ebbae35405b75b41481b34cfec53","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b4a8835d20266caa506a44d44d36ebc8ec5aec2a9c926f371458e0fea045fff6","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e14d9486565665d2ea54939cb9db23cbf3e05b7c089029d6d31a61b2009a07fe","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2482590","recipe_identity":"b7a902ca136f3ec5d27cc19f53ebab0cda0a26f7c173112f61551c1ba30e5b9b","loss":3.523836016022851,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"ac27689a553e2ab34cef40f0d2645527271d6b0851aef0cd2924fefde8840315","bytes":879},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"ff6c74a45d2a2887c1e3e726b97a3ab95325b61dee9b714f4e927051770b7b8e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"c3d481b23a25d5de5f5169a5466a317882e505dbf8c6206e0b0c06fb325bdd2e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"81020f995de67958a9e427b3d7cb35afecb54ffba7f0379bb5d12686e8995255","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"55387646839850168010c1969ce75b3693e54ae2a24056410ecf1727ad6b983c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"27f663199ae3835cb21e62d782c24ccbc0347ad14d9280af4a1b84363ffcc4a5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"a732abd9704f2ff5c9eec9eae09605d8dc8d9b85f28f44dda97400f10ccf4887","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"ae4b091a7bd8acd884279879cab2156951924793201241c15b10f9c7f9216431","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"3b2c833fec20cc21870662a44a9520b4b015e0ce4fc43db6c8b80bc046119edf","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"e900188c643446e1f07d99efd06c4675a74fb1d57711b5390f42223ae11b6874","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"ff9b2fe985b80ee27aeb3a6ef42c348efb0836e393369c2097c2d3798301099d","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"897f28263a4aca0d7fa0f9870733601363bb41bc4ac7053d0303d25a4ceced11","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"b619f4c16427befc7f332ff7b365930fbb05ca37620a6e6a80cc86cef28ad5e3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"c51649f13e08a87c2219da25c7a52eb47ae81383087255ab3646defb1ed0ab31","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"83a35511499b29397db0956d460b62d9d09b263c3a8177af080d25c91a859ef0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"84e02033cf27b77cdba6cdfb4234c3496721304da30694b7021810447a7a5d73","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"7e76ce12930fc3621723136bf50a91cd310a47f8c7ae1a3ff6204ab42b181748","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"6a69d26bc91c09bb1b53b3d31749001c2191924f219236f7403865d2f665b834","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"eb67d58d7544ca25b892ae2e1438fc881c85eccc9a8a947d95cd126531f2491b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"323eb44f26ee1950fa81bed0e4bb5086c35517259d2ef6e3afe2653eb81cf576","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1e7e011fed240fe4dd13832ff69555f42cc4a107b6f0acd559780c9b35821ad9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"8938709b5c287a565281513547099c97a0655e8f423a95ac2b0084a4cf9d2247","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"f377dfe8f5808febaf668ec0157591765b3600a1e1e5bbf3a8ef65c98718c4e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"ff2b8c4708ba1c92f53dc3ed1f70a7109d123dc1478d0fa4359aa85cb3b321a3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"2732732e7062590b547f10877b5c8b306008dcc70c3f4a621b324ad165cb9ca8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"5f26a0d65ea4929af329f8e22fafe5067f2aa6ac82e74e493b8a2c4db2351e3c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"f281b533cdfc9aa1836650ab687e0ade19f5f00260eab97491c4e82c417bd7dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"16e58a8da61f2ad686bcbe3822ba58f0be02027f65623636d7bd4bd089c5b2b8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"ab29c45c729a655b53b46c7de1cdd280c46bb616298ff5e30cb011196e80e9e2","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"45bfe2c0bd76499521380e3cb5225964e16ec4fe5cee62bc912d472dd6a0ec20","bytes":5261}},"requestSha256":"0079be14b850dd19a97588f7df0b0878985aa1e583dcf368dfeac2ebab37d5d3","metricsSha256":"7b6538f6d2f4a398e9f92a88106dd3e76f8b01d8acc96a8d5e563781f42ea899","continuationSha256":"8975eb8d430f7de4dc6a15921cec2d32278e9ddcb8a4c0dd417041ef7dbd0047","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1e7e011fed240fe4dd13832ff69555f42cc4a107b6f0acd559780c9b35821ad9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"8938709b5c287a565281513547099c97a0655e8f423a95ac2b0084a4cf9d2247","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"f377dfe8f5808febaf668ec0157591765b3600a1e1e5bbf3a8ef65c98718c4e2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"ff2b8c4708ba1c92f53dc3ed1f70a7109d123dc1478d0fa4359aa85cb3b321a3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"2732732e7062590b547f10877b5c8b306008dcc70c3f4a621b324ad165cb9ca8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"5f26a0d65ea4929af329f8e22fafe5067f2aa6ac82e74e493b8a2c4db2351e3c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"f281b533cdfc9aa1836650ab687e0ade19f5f00260eab97491c4e82c417bd7dd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"16e58a8da61f2ad686bcbe3822ba58f0be02027f65623636d7bd4bd089c5b2b8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"ab29c45c729a655b53b46c7de1cdd280c46bb616298ff5e30cb011196e80e9e2","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"f9175ce9b1b6765cae955061ab60bab912ac080d4daa97bd37e32e631773f178","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d78484fbdbbfed0fce0a944a30c0725e37d82d083d81e98013c0f3e6b1754891","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b7a902ca136f3ec5d27cc19f53ebab0cda0a26f7c173112f61551c1ba30e5b9b","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2482591","recipe_identity":"8ebf89405196743b4d09cf2bba223566fa8ae360cd2447f1b45d37eb31fcb2aa","loss":3.53144581791852,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"a3eaa9df3c847c0513c2f8457e224b115f31dc91d6d887ad67f56e4db25b1d4d","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"eb4bdd75886de724bd91d83da56a8887fba3202ec608574a51a86032e2ff7f2a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"98d1d34cc5ac09b995900b61124155ca7823ca2765701c883166a2acaa0c9b70","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"058a2072cdc56772d2b6803f0cdc75805942d252c5201dce9d0e78c1af841c93","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"c6f15de8081b9f3afb8cfd055aeac718fea850b47db1436190cf70df07eca9ff","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"6a1068f77e7c857c87d54e98609e1c64c0d014484114e0b3587e61786ea7dcdd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"3185afc05c2e7b213db31633316825115f3b31740508e9b4404d820f54da715b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"deae4370bd9b0eb2e5b57a5580988aae6ddf818d8d4dd7d88c0e67350b4cb0bd","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"45e97bd09d3c4da4f7dedfa816dfe6678820989338c3056cf5ef54a6d7822f62","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"be6adbcdc94fe0243edd36303e3bdc6aabff95f57770c604bb6dbf1661398c3e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"37783116752fe957e44ed9f188d427c84fbe64e6f99624ab00f95e132dae3005","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"fb0cc617a4c251094dabdc3a194875d3b20081273dc3c91940974a762fdfbb97","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"10870c365e2a0d22a01ad80ea819c588c7b3e7291be876205c8a92f9bffa91f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"cc476cddf4e393d756be7e2cfb20f9074658d90a520329975669cae502f51613","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"7e7c2ed1a50dfe4f01d00c10d3cd01e9705f1f48d14b90891391642c4308d99b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"31c6ab03885e091417b44b0f79dd23e22484f9324cb419d4ab34e970d24cf852","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"c0c6d7c383a0eed78aed3b25c7041ed61ea50c6fd12176f120a0eb35168a0822","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"a2bc20578a679db2fc265ae901e43c762a75861a158e0439c1915a6fe2df1264","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"84405428e11130475bd209b98ca01326529ca4afa6df150d2523956f2c1d1135","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"409659315c8bf016e260e6bcc620a326cd605cf8e3161f4d393f7b7f82e990af","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"9b50335c78a0dc2624e460cb4721925161b5a336ad453f8e95ead676bf48d065","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"d44630f016fdfd1749312d942ac032101087550796f19b16f521b2b086c20913","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"a36eef239b6d41665aa1df58fa87dec418b5560818d54ffd122d48b51da5f735","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"e15c8f4628bfd4f642bd615412a053129bf471b5dca164a176ac70c8a06d36c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"fb4ba1d384eb5cb397b2d8a96f1c9ddc3282d2e955c16a355586485bde433e6b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"c8ec679e8e813e099d5e66b3b4e1a9b3267b13b9853bd224aeb54264e458e525","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"2899c1e45f25147e1493e3c1f920152e830a9a41611fc3e2bb00f8f2453552d3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"e3b097e666b3912ef929dc4b3824a2bd8cf1cc2f88ab687e78d7e7d866347bc4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"febefd6761325214893a6a93f53e39bbe8ad1fb4fc2dcb4f339d3576e4be0357","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"7d593ead82070e29a828cdb91ffe4ae58373411a67f6674b46839d6c6c446bb7","bytes":5343}},"requestSha256":"5987ce5b7ff093b28813e4f90b7ea7b8c511732a83b87049ae84dc25e31399f7","metricsSha256":"f115307e415f6c310f45fecb0b03dd44dfb2facfe0eb96046f774cbe0a6736e1","continuationSha256":"3bc4cbaf425301d75475620a6cd440c4b7c19c6eab17217e81b8e88ad7070431","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"9b50335c78a0dc2624e460cb4721925161b5a336ad453f8e95ead676bf48d065","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"d44630f016fdfd1749312d942ac032101087550796f19b16f521b2b086c20913","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"a36eef239b6d41665aa1df58fa87dec418b5560818d54ffd122d48b51da5f735","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"e15c8f4628bfd4f642bd615412a053129bf471b5dca164a176ac70c8a06d36c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"fb4ba1d384eb5cb397b2d8a96f1c9ddc3282d2e955c16a355586485bde433e6b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"c8ec679e8e813e099d5e66b3b4e1a9b3267b13b9853bd224aeb54264e458e525","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"2899c1e45f25147e1493e3c1f920152e830a9a41611fc3e2bb00f8f2453552d3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"e3b097e666b3912ef929dc4b3824a2bd8cf1cc2f88ab687e78d7e7d866347bc4","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"febefd6761325214893a6a93f53e39bbe8ad1fb4fc2dcb4f339d3576e4be0357","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"07f82924041d6167db0d149e641de75a4b5d44ab910c21d038b1b6b472f3dbf5","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"726114c13eb4db67a0b450a5336e5e1eb52e3ce013b73a721033f9921c2c9a19","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8ebf89405196743b4d09cf2bba223566fa8ae360cd2447f1b45d37eb31fcb2aa","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2482592","recipe_identity":"21f1491540b88f787c4189e1e3696c9b13e6d281a771ed884926fc595670357d","loss":3.5409533873909926,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"70deddc9d57a6331af3bbbadb2775770e032614cd43f9ffe924ed302a201eed8","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"30f3d707d4cb900c9eb6ef7cdce72f8581092bee5d1db5e6f614bec592964962","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"1fd70076ad4092bcb3149aee1934878404caba04040b57c32a1d422c12035afe","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"6fe8c1e4fcbedb1e83dade0675368fcc3c477aa4bbb0929dfd4417d75a84582f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"a0a4208070e53b236d32ac77f0ed09f11c7b676c8351f3ad789bbdcbae52f518","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"197be1cc27e054f8f7b2f8e231209205d1d55d854146ec3fd6e685abc72d0f48","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"ac0483f5da30929b612b33b7c75ab4a61d29cb314c1d5c3afae51079066baa11","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"6d251c90a4b58743d80b87531577c0de848e8125bfdbf1c9ac4da791a2457db9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"c27a4b0999789851534c32f4e062e985cc577990124a765b79b3ca1abde5ee00","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"8a95a118463063ceab6e806eb681af66e65dcc950e2e6aae18c31c2c73aba1bc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"2b39f90d0471a38aa4bb8a6f2414a6dc97c2c8bb44d241ec7d0d865c1ec3b8e5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"dd07b436aaff0c11bc21dba45ff06ab8e927d4b904e7acbbb7679bdcdce1abe9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"39d570a4c24ecef81d3f33cbc2996e182a458738d97ffc7c99e81dadcc770461","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"161a6be4dab2500aff55f9216c6298be3b8ca9896fdc96d66427e1753ea03665","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"8110862e1e1c039d3c8405abd8ed78c9652b88a90606e526fb9973c59b159b26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"d080d9dfa4a40a03665c1dab7d1bab0b17485d63358e686d42cd893986a88d1e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"b7713938873f68d2325f67a807141a36a646cb61cfdc680544058b8dfb9dee90","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"c6c555ee4c42256e635601ecf4acd787ba12b3200ae58168a2f2144e8ca5ebc6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"099caeadb630402bab7629d482acb28f4ba3d9136137cfff78e52120a0b05e97","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"f1b19aa7159e91392620d6f3b86529675acae14909d41409afeabc68afdaec3c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c03974001655591a2894bcaeedae3fc452d4e0bf8c27e0dcf5911d99a04709ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"7415222b54b0b887667d4791284cdaffc6d2b424154812dd7e19df6b85f6dcff","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"59e1714401ebe35737ba0cac3cf3d22ac949fe68d5e2923228d930062949856d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"42fd891d883dc0a08634d9744dd600b1084902804e534f52f1a28ece05b61cee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"c2f5921bdc0bb1090fe101aaa4cd9982184929d82f153f5a3ccebb30b065dc10","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"a80c8eb2b0c8a1130959f036b890f6fc7222eb8efbfbbd878871272ba48155cc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"571c66926ec321f475c9160a74e170ebefb44aafad6b45a407d554359b9cb10f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"3e56a30e786063ae2a7e71bbe2d455ad02aade739b13e110cf639f3f3776875b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"8df1578215e0fc5eb20c78c5b5972762fcb23221edc70de9d043fe03233f9781","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"e9385095a83675b122ea040fa5329502914bd57238883bc16299ebd8790169ba","bytes":5314}},"requestSha256":"7cbdc81dc7094d8793db05c4e69ea4bcafcd31e91b3adc552f3734428e834d00","metricsSha256":"8d1f8a76c8106592ca6fb0d9afc4cfd4856846cd9d5d220093900a6852ab855c","continuationSha256":"fa7236baf85a6655f42bb279257cb41855ca3772df9f893d868fa88254edcdfe","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c03974001655591a2894bcaeedae3fc452d4e0bf8c27e0dcf5911d99a04709ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"7415222b54b0b887667d4791284cdaffc6d2b424154812dd7e19df6b85f6dcff","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"59e1714401ebe35737ba0cac3cf3d22ac949fe68d5e2923228d930062949856d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"42fd891d883dc0a08634d9744dd600b1084902804e534f52f1a28ece05b61cee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"c2f5921bdc0bb1090fe101aaa4cd9982184929d82f153f5a3ccebb30b065dc10","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"a80c8eb2b0c8a1130959f036b890f6fc7222eb8efbfbbd878871272ba48155cc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"571c66926ec321f475c9160a74e170ebefb44aafad6b45a407d554359b9cb10f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"3e56a30e786063ae2a7e71bbe2d455ad02aade739b13e110cf639f3f3776875b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"8df1578215e0fc5eb20c78c5b5972762fcb23221edc70de9d043fe03233f9781","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"ee4794c8242b64771367c47dcabf62d59a9dacd3090e7a5c4b64425d3930e106","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d42f2c93ff946079f0b6aaec500543cccf45b203a75ab839f259150cde1f39d7","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"21f1491540b88f787c4189e1e3696c9b13e6d281a771ed884926fc595670357d","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2482593","recipe_identity":"ecdceb353dee6714840128e40f5e4f54fa6e974af82c66fb2c5ac2223214f1d6","loss":3.537012955573432,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"1c6a29b641009fb877b25617eec38638077fdb4424f56f781dc778d76e80c71e","bytes":881},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"e7b96803786758975bc373c240242043a7cce9afaceb610d0db9886c70e5123f","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"8b96e0db7723d7ec381f81e42af4f2c9c6971be6b6ddf5d5cd078b545d47cb41","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"5ce10622b25b732dc091fac5189c7c98645e9f6b16d0fb005ecbd965134dee47","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"8b4cae8bd4ab8fded205c7ae15e82dc96a03247956bb1f7b79b9d5901564fa46","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"2b7e6d130d46f1f8afed4b97a656820eb4a933d1e84cea3b9d6b74270eea2478","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"aa9f8363da750a59dececc7dd771c2228a69cd22d402fdedbb826a8fe2afd4e5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"c11630619240d2ee0e860bc70808c2ecda118a10962330995639d1ac1889f436","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"92363936e88c68bf039fa1808ba716c9c9f213ea09c88000acf351ca9860ed86","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"9419a5e0c3adcf4f106b12bbfe0428fd5b112c214520c6d5edcaf2d24975c2dc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"2428c976f9d76e792f79810770bd094845d5615e651536178b7db0b11daf1a94","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"a937fb53980bbf4ea33b7d4e1a9577f933edee493df07156b2c2d40ead7272dc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"9be1107a417a96cbeb88eac5bcada0df4bab328d6b83b2972edb064c8729dafa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"152e14e89c7904eb0dac627d97832db159a043ff26661bd8e1a0a4ea5c5756ce","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"48e8d253c52e4b0b316e5c8718bbd28be83dff15860e9b42297c6a1f0c471a9d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"75e502b70e5e09cf18c1a77465ce415876a1939ae4155e04fb53cdca87a58604","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"952eebc4989e2311c523f829e90e668264ed3d63660ad6a4ef4f1bf88f5d0c5b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"0686a869f28cbea83c0f880c61db5eb98404cda23ebe36b068386e086252a784","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"15c8a8a05233c69475d102145f72e91fddfc1b2d497ec639e24e6a8d526f800f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"d412cc4114ee13b95e39ee4af1c0da6a4e4451d288ce8f3175d2b12087ab3632","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"642f155005e91243386e8d0d9a9c0090d91ae1412ff76144c1efc8bbf5462879","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"4d167a223af2fc824a6e0aa9ee9504d9f3f4ca397ffb5aeeb2d5d8d0f4ac8812","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"58520601848da0ac41757c22b846e0d0ea961823fa8138e9a729465627e316f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"bb557ebb026292bf287597c2f6ad4f2767796fe4a3a9d775435766b5c8f4635e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"a7d270b19dff0f9ee33ce79733f470dc4f05e3620ba672d8da6a96a74c17585e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"33ef3f6ada49e54853af23bd3cda0803eae6713fd24af1f521134e1af8ad871f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"61647504ae24fe20b733d00e0572acf84163e9391cd23a98277e569e44bebd54","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"ff0db2e39d9575d6b1eaed6936c57a72ef2cc80ec682c04cfaaa9b3750c260be","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"02308f19288ffebb082f43e60d2ee97fe745a0cfbfde2ca7968d13888620af9e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"e64854e8350dc27a295a7d45699429c5f41773eda0d75738fc2db6afbd3b5e4f","bytes":5339}},"requestSha256":"0ede9f29f4a4afb0735d9e33bc96266e3d8c5bbcdfd9c61e6cf2c419ee349fff","metricsSha256":"32e0bdeb08fe9d2e599368de581f1fd3c8f137842d38c08b39f34e0662cf7ac2","continuationSha256":"94c6e5f66c6d787a7c1d0ba1008d0dbb92b2122d1774aa4c87ac4f179ad23aed","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"642f155005e91243386e8d0d9a9c0090d91ae1412ff76144c1efc8bbf5462879","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"4d167a223af2fc824a6e0aa9ee9504d9f3f4ca397ffb5aeeb2d5d8d0f4ac8812","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"58520601848da0ac41757c22b846e0d0ea961823fa8138e9a729465627e316f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"bb557ebb026292bf287597c2f6ad4f2767796fe4a3a9d775435766b5c8f4635e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"a7d270b19dff0f9ee33ce79733f470dc4f05e3620ba672d8da6a96a74c17585e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"33ef3f6ada49e54853af23bd3cda0803eae6713fd24af1f521134e1af8ad871f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"61647504ae24fe20b733d00e0572acf84163e9391cd23a98277e569e44bebd54","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"ff0db2e39d9575d6b1eaed6936c57a72ef2cc80ec682c04cfaaa9b3750c260be","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"02308f19288ffebb082f43e60d2ee97fe745a0cfbfde2ca7968d13888620af9e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"b7d5473fe9a5ce81fa76753f691c52cdb829a2be90db4fe4a606ccc6f6c2ce6c","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ffcf85665df13ce512af4d02b8a298aee9a77c498a8d08127e34adfe487f1330","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ecdceb353dee6714840128e40f5e4f54fa6e974af82c66fb2c5ac2223214f1d6","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2482594","recipe_identity":"5fc7942e76a196423554e6b438618b4f9ba58b246580b99d2cafbd3d5ec31f4c","loss":3.5427450870510637,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"f2d294c8484b5c5e27122fef367902202a87c9648efc72d416e77246d56ba338","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"a5210208b686285ebb1d7703fa80b280b7441292282b6e14881472739907e373","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"2b8cd524d28b93e38f38c1c22bfc3fe9f73067ecc1feefa79e9e26ab3ca7cb86","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"a494713528b4d486c9cfabaaf6bee15167ff773541a861b245aa861f8d169a94","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"7d1014b28302bcc044159f6cf29d85243bd0934b39f002bdb333f5f4a69bc4c8","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"57b4f59b43b09741d84e6d3c14796f81b1b8588d8f7cbc50a128d7a11ae6c856","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"f47ef1920d5dd1cac394a65212ec05e243d65e17a3ce9f3425e92553b02a00d7","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"6df020e85e693cf3af712a3bb986017603d77c8cfbf83c95e910a67c3ea86808","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"d4204b559851bddf432b69f4a587ca0e9c2ba9c5bc511cbf6fb069b7a73058fa","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"dfa63b638e6c360ccade8d748435655ee045c002ef941a331ebdbc25df011d59","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"c7642586f21246ca593c94300a24d4240ac0067d96e6b3089ef390f6379c753a","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"a437f0ddd35813d7fe33c5133968af05620ea38217364fb56815781bd5729f7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"8763601341437fe8e01fccf8ae1ff0dbedee7698537bed0b416803be50646d02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"54401e5ec64f26fe0fc7440af95a7315797b2665c0f77ac26c034e91599b2f1e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"5a2c4bfc6e559369f91d35949a0a80a5c758f9dee4ae7d31b7e493ec8dd76e1d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"16fe379c1a072681db46113f7e8ea61e8b1d5975ef3202fa8efd9772275e0462","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"6cdca0594e2005d48f7759970036dbcc9927e61aedf28531b2801d96faa65fcc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"2ef1da29bd6207e50f52305579965dd56d41e520d09cf0d007b35a47045b32ff","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"77092a37177297ec7feb8a1d169c75be672fa522f948cfaf4c67ee79ee248523","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"8d1d15b299c5c2ec8409f151d68b984c846d72b7e4482137944e05fdb988ad83","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"305a46ac5bca28636f31c50e20a4ac51c9d240b2e35097c9b83b4011c2859ebd","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"36fb2c069e1d3b22ddf8401098f41da7b026b6547634424ebeca576cd0cd62c2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"ede1b2c2df0f43f3db7e1ce2fb01f204d995b45b74b8b84375a8dec790f63d51","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"116f3b8013651e1b9b2bcb3e292abfd2525bd1313ea36f69a9ed6dc45687cfdc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"1a188ad1a47cb0b381de2ba0f87d5b7f7a1272e260e98e70c175280feb278be8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"54878f7b054db9362f78dc8eae3da9dd1c547f792befc241bb86c227ff2d792b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"0cc132ef9306ccaa591adb92c63d844e33654dda4d9a557360f86fb10df9263c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"27bc78dc71efd8d5b44692728044b943dd84d79de8b05f45797749c1653ddfca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"c49984b878313a2087de127a3738950fb813623ea4c28cd740f1b693c6292d3e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"7eb4da72c41e446122370ac25dbb69a7529de7588fdbb84cbf5fe4b2291c8eb5","bytes":5256}},"requestSha256":"7684c5b51c70154f12f1bfb24891e7dc727a23448812267ddb4ab9738ec96e30","metricsSha256":"315837dae5f165808b3100ffca2187a01db97e745f0813a2224e53cb650ddf65","continuationSha256":"9732a9594127a84700fcaef78a4e9be7af1c34b8c3c294a49a3ce0650127355a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"305a46ac5bca28636f31c50e20a4ac51c9d240b2e35097c9b83b4011c2859ebd","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"36fb2c069e1d3b22ddf8401098f41da7b026b6547634424ebeca576cd0cd62c2","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"ede1b2c2df0f43f3db7e1ce2fb01f204d995b45b74b8b84375a8dec790f63d51","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"116f3b8013651e1b9b2bcb3e292abfd2525bd1313ea36f69a9ed6dc45687cfdc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"1a188ad1a47cb0b381de2ba0f87d5b7f7a1272e260e98e70c175280feb278be8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"54878f7b054db9362f78dc8eae3da9dd1c547f792befc241bb86c227ff2d792b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"0cc132ef9306ccaa591adb92c63d844e33654dda4d9a557360f86fb10df9263c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"27bc78dc71efd8d5b44692728044b943dd84d79de8b05f45797749c1653ddfca","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"c49984b878313a2087de127a3738950fb813623ea4c28cd740f1b693c6292d3e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"216bdb52e8ca7127f726e3627eceec8b3292774eae6285a19a5ad01850412cea","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bbe93010ec1db7eeb1b69352d308c493b4b2686a6ef81bd08faa0df9e28006af","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5fc7942e76a196423554e6b438618b4f9ba58b246580b99d2cafbd3d5ec31f4c","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2482595","recipe_identity":"95a69f33712f12107949a8e3f873b9a7e15ba4b4050b4c67d166f4930e0add42","loss":3.4219033785359185,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"2019c5417c85cb27470abdea3e8e72ee95901fe200848fc7ff88e98f218902e4","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"7756e546b2dc6cdc0007bf5aa0edc0741831c9c6a98c9b03dd7c6e72a64a6314","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"63359dc612cac61fdd4d2990c8e9c092c2e9c80c1c33676c8fff538c3dca9824","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"565fca81cad1f8fc8c3af94bd41f7d16aed678c93b8c6ee6e1aa907c86722b60","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"2791a2696c5ea5a7d555e5224e9d709b9e66b6ed3557ba629a8e789445e0dbc4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"0e563876483cd5ac046881358d82cf4e782603b662ee34e94b5abc93b5af481d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"0de24cbd5b2f3f1c1b65ae98b3085b29f21a5a5c9cc4ea3a0eff039835ebc4e9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"a3f2782f53f477b018320b83bcc2bff664dbd9eb9783bbc8d991ab995622950f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"9c91b845dc696637c640413f79ed5b31dd289655fe62a75e7b0905e7831f8553","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"bbb7219ce5454a723721c1bec10a5a3be4adaaa41851d75d379b7e36687102a4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"f3f99fd78885a2073d125f41f6af7ae02ed570c9a8f86f883e9961ead5b3b85f","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"223f19b3e2318725b7f1af30b2c313d26855f99ed9afa3992071e6db27d4cea2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"1efef8666699c0fd75b8346dbb8647ad67d0eb2e305a8c0c976afea22d221f20","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"c4aa52db8e3937a3c91f2bc0e4d04f68c08271dfce47be3e4dadd657ae0cd8cc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"a972981341aaeddd5ca25c2fdc1f18f82801195fb9d997377b511fca78ea1ec0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"4524f3b2dde442407caf340d37402a0b873f057d87c6acbafba01f1efa83a1c9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"41c0537c5789338edda9073327882a2cd21bc3e3f24405b2cd30ed9093dfebdc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"ee37090395388f9ce46a0d9180418a0685025aa9cdc86fc052e047a3f2786b67","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"253f7f8e946a678c25cbf7bdf0f461b67185a8b9a8da33f6cf0d6389c5f7a64f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"ba0cc4f4584f7478f95d1645aa08d434949219cb80f95936a7dc8bb292d874b9","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"621118079a58550a301710de97806f40ea129291b942e606b9fdd69b1bde2390","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"37d485fcee63e3f3994365ba716c3c7a0ece9799bc2915dd43ba7d5107479d17","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"d7f76a287c46cca73813fa01401c5c82bda9de1801df6d8a8c4deff08a9a4506","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"76958fd0b9aa24f64d9c900a47439bfc2f24ed22914175b99d85cfadf4e754c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"f4c502eeb9d3088cc36a7918faff6f27552c38f9807ee9872aa332290ed0f331","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"1895bc355e9e0e53ba5f22c929c51a9da933204ac89d080f278e5a9fe5eff0e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"2f163794082e7baa3f3658e3b635812c4ab7e27c80b9b93237faec86f574650a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"c03fbc2e9eac4f9ddbb8ace553efad744e0782238138fc440866a5f11d03eb2b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"41ef33ca5a8789632decce90d4f7ab900212137969e0a3a288543850dbce6e98","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"fa6db96305607e5f8e451879b16ecf087d9bbdabf67f4d1d8dde79232aeb7cf9","bytes":5261}},"requestSha256":"718a4abb9a68f2692a420736769b329129ea720e3d6505c02917111caaad919b","metricsSha256":"37b235a3184e12f23222d5126810e02078c4d82cc567c7c9c78ddbae113ca00a","continuationSha256":"b1b256056c852049baf797f2853447c3f08949e7e4c461fd65a038c4170dd6f7","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"621118079a58550a301710de97806f40ea129291b942e606b9fdd69b1bde2390","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"37d485fcee63e3f3994365ba716c3c7a0ece9799bc2915dd43ba7d5107479d17","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"d7f76a287c46cca73813fa01401c5c82bda9de1801df6d8a8c4deff08a9a4506","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"76958fd0b9aa24f64d9c900a47439bfc2f24ed22914175b99d85cfadf4e754c1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"f4c502eeb9d3088cc36a7918faff6f27552c38f9807ee9872aa332290ed0f331","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"1895bc355e9e0e53ba5f22c929c51a9da933204ac89d080f278e5a9fe5eff0e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"2f163794082e7baa3f3658e3b635812c4ab7e27c80b9b93237faec86f574650a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"c03fbc2e9eac4f9ddbb8ace553efad744e0782238138fc440866a5f11d03eb2b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"41ef33ca5a8789632decce90d4f7ab900212137969e0a3a288543850dbce6e98","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"c1616f5091d9d60179198ddbfd11d997c06d7418d4c225b7a0635fe95454215f","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5a456ccebca83a3f9e12281ebd63334c1406b131d943e2a788ea5de9c8fa0466","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"95a69f33712f12107949a8e3f873b9a7e15ba4b4050b4c67d166f4930e0add42","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2482596","recipe_identity":"ec11577a5cc4f74475bffa54661304f8c8d9d21c4567a0001809463d8d892e39","loss":3.426957445863731,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"4b1e2b336fce136cba70b81324f6f026d0ec459eb780f549d27c224bfe923ae0","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"f7b0528d4f647d5dacbdea9cfb92a0d5203e49331de0302c9b520e66d6f72e4e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"1f64cf3a57ffe904c9ad6260fcd87e82e6f9c53c7898850931b0eb8efe019a6f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"c99732d04af319a72091b8d2fdf448d8bced5771c487669093fb3d0a0621ab45","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"4b37a010e91cc1674974d3e99b281815cf989dfbfa6ea5a35eda7b1fb98c79c4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"b1104be30985a7f2fd8d40cfae47dc919ccebd79108c4106bb8d5f203afdca2b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"6e840fb6114d660d970e7e069bb47926c8941dd862c810778fead501d9548087","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"90c3b29ccbc6005ba47447458b0782ef851064a767c59220a940858efd511e9d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"19605ce2fad943ec8e9d4a56a347130c358c62c32f699d4cbd661d5b0cc43579","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"0e5a10bcb33e18cb83cd20b96a1eb1f0d0780b8b1415c3906f75e7dae2705d9c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"702d022415972bebf43a38fc2e2da44bbe71b0e6b0baa4d7b55d0ca13e968457","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"5fbe452069d93b1948deccb13b4049911165130db85d9a0470e0489dad455950","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"d9ee9b166d3b692ce81ac828fa2c9caac3bf664e1615eb4c103b9997d4f0ce6c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"f2dec92ae5e392439e27b672d05383cd7ee8e83855d1b5ff4234c95b3ad0139b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"4a759b2bdd04c2c8421682d7b982bf5034c932af58c8480c49333994f1bc76a0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"a14bf1b6d8a7e3455cf9d3847d48718ff4cdca053af4fa86479d1a73f642e115","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"0e8c54dc3a3b0e386b4abc3ef3b29cf3aa58db21ab5e70f4d30b08f95bb0e89e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"b5137b539965a6a8474d1cd0a90712715eea8ba735a6ea67e85f3a4442a7b5fd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"3fc3b9113ee6ab9175fb95dffc5928181583d4f0ae020de51f0a0c937e816bd8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"ef5f9023fdd546701385b50bc19914ab71bc3e47f071c1fb536dfbe99f8c19ae","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"b039cce8d74aff65d1bba0052b640625f4652dd6c1c8d7b157a4fc16261e48af","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"595920d0e91bc7dc32a8e7cf48cd0e721073810de777160e592dd82bc3fff735","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"fb164aaf440b9a136b03dbe629bc1b774d2d307c3e735b6867daf5528e637c30","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"8fe060b7c18d0e2ac902516a62b427ea0cb4feebb73d09d37e80ad5a4f8ca7b6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"a7d52a125263e58838cb8c8934e3e6d9a8747772047670b04214f0ac2bd5e81e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"f9e40adedd8d43d553e4ee07a6f4c674a9052879701d50debcc3c37a58831b74","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"06a821b508465095fc966ca3ab96cc1ed410729154dfa52486aa7a716c8b98c0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"e8a6c589911a301d6291890f33264b9f07c1401bec8d5f55c8793b92503e3946","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"880b54cf3b3764d940bfcf7bcfdf0ca14d76ae740f2a9a48fe856205bc12b34b","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"c6d22b81f9b6c30faa5e854cf53cbc0c81cb07c5b46fedbfa062d61b2f5300ca","bytes":5342}},"requestSha256":"e665f958527fc15c7e01c8f1f3c48d946c25cc0858a70053f345296488e375ed","metricsSha256":"d0145d1c39e618c2763f2854f0e5fe1f4e9e31ceefb8cef6fcaf9ab48082159e","continuationSha256":"aa8a4c76f1fc31843711df78befdb21362bd243f8f6f59be77236f8234e8f59e","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"b039cce8d74aff65d1bba0052b640625f4652dd6c1c8d7b157a4fc16261e48af","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"595920d0e91bc7dc32a8e7cf48cd0e721073810de777160e592dd82bc3fff735","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"fb164aaf440b9a136b03dbe629bc1b774d2d307c3e735b6867daf5528e637c30","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"8fe060b7c18d0e2ac902516a62b427ea0cb4feebb73d09d37e80ad5a4f8ca7b6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"a7d52a125263e58838cb8c8934e3e6d9a8747772047670b04214f0ac2bd5e81e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"f9e40adedd8d43d553e4ee07a6f4c674a9052879701d50debcc3c37a58831b74","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"06a821b508465095fc966ca3ab96cc1ed410729154dfa52486aa7a716c8b98c0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"e8a6c589911a301d6291890f33264b9f07c1401bec8d5f55c8793b92503e3946","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"880b54cf3b3764d940bfcf7bcfdf0ca14d76ae740f2a9a48fe856205bc12b34b","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"d70ab0b0defd16b624a4ff982e86c08c258603b595b206e3e85370dc1d1595a6","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d48039310e2d4e202c0eb5b27c4b65aac3f316fb26be1cdbe4439bb97a88de28","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ec11577a5cc4f74475bffa54661304f8c8d9d21c4567a0001809463d8d892e39","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2482597","recipe_identity":"b8c41d9534f8bce0bdda4614fa428ab5cdb9dc03b25dc4399eeb70aedc252dd7","loss":3.4174817738686873,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"46cd6f5a2e31b63a0dda80c665a68ba1b8a2e92d64837f231c51ab5472f870aa","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"70b55ae22dfc74b91ad1f851f796f166b585bed790682d7699354a5911a59737","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"bdc6a47267ecc9fb631dfafc374959fd3335d1c67eac3f75baecc9a0c7609d00","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"be6099e69b7df205ac0a946e459bc7b71d00b5a9550e25cb307b9fa7f1148f93","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"bc305a02cea535addade5aa4b0578f541722b15b12f70a6ca1c9316a68d5dd8c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"2940596b51b074cdafac265b12c198b71b18abedefcbd05640e43f809ad78bc0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"fa7636e4fa778d2b3782d69b08f43cc5d6b774d6583eef37d94787fb513d2029","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"6563d1c57cf8138ed663751bd18fbea82d2a7a521d19b52a981decc71df4362f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"89e53cb5d248034646a09220d566d4e2f4b1e72fc5a234d4ac00f68010545deb","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"972dfc9e83441a6b4be15b38f070d2745de6e135c66e0972b78d6938edde66ae","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"6eeb94dac155b9aea8c422a0e6e3951f84171a5a3743b0077070e7a345308dd0","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"633b7e11695be37f8289f1ba94df4c0a3b650a1d701584be4d0aa98266a35954","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"080e43eae6a77925b0e5659b517a7eae7159696afacd56cc8e5d09a30d108373","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"dd2052506078b5889ea481ecc4411d34a3edd4381d9d5abc1431b5ffd8763033","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"9ba2c8cd899551de5fb3e9852496ff1283e34ee62833518c85d1a6237a1de49e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"c95bd14a1101b77d2d6a9ce0c332775f7597b03deb0b2e90c7c6ea7094447bdd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"c3b1a4a1d793e415d0b0b508305979feb46d6fd4c23a58b1654c9cb7b6e68a76","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"e29c17585e68eb4b13963a9542dae6202afbbaa901c2ca9b0dba7132d74cc183","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"0bc4a88912b4c36b26af38416bd0c76eceaf8a01196d3b82673c01ad625c5031","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"b78004e475cf6a3e23ddb7978220abcc176c757b8e6ffa201e39bac843e7390c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"5ce2c3c00466941c36b424ca9e8113ed04bb017e8eca62ca4054e06d07fb4e70","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"7f5f5b5ae2b65b06a87edb93222ac8c35a65bbd69810fc3d704191e857d8dcbd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"dabac010ad0a9c3ee0ded0d52228549683f2667fbced97328819d5c41fc95e5c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"dd2f1f31622a3fd7d7730be327cf474c573a475cf3eaff99f46c15e3c48d67a9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"8bbb0b714c75bc92a3ededa89c89b8dac99f0dd57ea66f1fd8d3a985de80345d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"7348912896e7e5d1905c11d5cf16abd60926b88b3d1defed4306d54c06fdfa99","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"9100bfb7c4ebce9b6f51bbfc9751f6273cb0f5968c1cb33c5a1b2550be690657","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"f0035f45f3fdc3434b323180ecc277ca5e1a0d3daa598a4177e7e6796057e1b9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"8195f534abb815b7fe97ac80bfb73daa5de5c20c3e937eab5b8fc59a28c17fb4","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"292610d70e4a9a3f0711bf5d7b16d45a64a9ebeef317748c874caca53bc386b9","bytes":5336}},"requestSha256":"78ad52c03a5a9bc2d59de79a5f3b9d04ea6fc97e94c73f4eee61ea7c2cbac847","metricsSha256":"b6ffbde7c03b0b6c737c24923afe16079462d59c69069fe2c0b23a53dbd450c7","continuationSha256":"7b9b6bb9823cb089733a21812d90052978cfc6cd17016d160aa5921143eb21c6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"5ce2c3c00466941c36b424ca9e8113ed04bb017e8eca62ca4054e06d07fb4e70","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"7f5f5b5ae2b65b06a87edb93222ac8c35a65bbd69810fc3d704191e857d8dcbd","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"dabac010ad0a9c3ee0ded0d52228549683f2667fbced97328819d5c41fc95e5c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"dd2f1f31622a3fd7d7730be327cf474c573a475cf3eaff99f46c15e3c48d67a9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"8bbb0b714c75bc92a3ededa89c89b8dac99f0dd57ea66f1fd8d3a985de80345d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"7348912896e7e5d1905c11d5cf16abd60926b88b3d1defed4306d54c06fdfa99","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"9100bfb7c4ebce9b6f51bbfc9751f6273cb0f5968c1cb33c5a1b2550be690657","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"f0035f45f3fdc3434b323180ecc277ca5e1a0d3daa598a4177e7e6796057e1b9","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"8195f534abb815b7fe97ac80bfb73daa5de5c20c3e937eab5b8fc59a28c17fb4","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"a5e6ce0d117a9994b9bdac6e4a265150a63ad985b8b5658018fd98346f0afbc5","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"58c2b208deb6e68c14ae88380404eb05da069cb948be94254ece10179f323772","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b8c41d9534f8bce0bdda4614fa428ab5cdb9dc03b25dc4399eeb70aedc252dd7","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2482598","recipe_identity":"4c0567ca492a2d7b60d1c8fbcb75abeab397537c2f8a26a15a5019da7c33da5e","loss":3.4369610570896363,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"6e5fae69cb95faeb3d87a1a0a5f523bd655c2b375da48510044b28305c9d5301","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"a54658e107078b908ab5261a19655ad85ede03f4bcf562915f09773d226c28bc","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"24320e7307edcc5da489127a888722fac8c4a602f3d09f9d8616ceb04a6ffd42","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"79500811a76fc976e874e198662fbb3d0b5fd7347eca29ffc694dee09ea0897b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"1c47a26c857396cd99ba5178bea4a5909df9134557bfdde3abb28ba650e9c6c9","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"d0ea742e347423b1b976d2823fe81e791e47c9e707b287a6f2ca8ad35d352250","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"f268e29e52343416ac4e9b31a0a1f6b6a632887620c9ae2efec54a7666055e8d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"932ecc1d674b9b9a629e26596c10033f4b023bcc6b2ab990bac3d23dcaeff1f4","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"f4f10dadd35e7c6106ee13585d322f1902fbbe34ac0c5e4a5d6b991918039256","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"2b085a5b7c03d04f0ff242a204584a1277ccb58da001cdc2ff3d2c520c93fcd1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"06730acabfa75cc3950ef341d67fe8fbde9eeb6ff6642192a2e1e56439d4ce1b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"d032e6a881a8d9c917410efb53ef35d996d10945e1d44aacbaa46565a7c962e0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"2e855d4bf2b738e00abd66f2864451f6e53ef851043d16070db42ee55f30e8cc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"a46128e26cb29bdfbab762fb3516d595f77a620957fffd0b9a54349d14fde0e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"a0205b03250cb7152f1905b1359e3554750b96d8c5121dd7606fa33b5b154c55","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"d08d8a5ea5645e536b10d1ca94afc848bb7834a008ed62ad227bbfb9ce134ed5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"2dc5780bda3521f1ced2d565123ff0b435267f85965b63819c9f286902703a10","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"53cfd3234088d003c435b4c13250c2b64b9f08e00b099cb12010e9b5b1fc0b3d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"6047ffe97775fb5903bb984423f9b0a346315fc386ae0be1f6ed9d6eecc61c32","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"649e62cbd29338dd597bd4fef33521f676dd522621a778441212c2ea091bf9d5","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"3499cba5c2f9330d9fb4fa31df05ef692e2fe2bc5fe30411af8606f7730d6a26","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"b8c2f65a4effc4e48d4c3299f6fb54dc0d750dfddab4f8badf5e25bd432cf721","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"9e8cd05f0eb65d7e0968ea431247cc4542956f4dd21835724861213dcce033a8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"90d9f01cda12261fd69ac45fea2131c3459c10a8ffc5a665c1222daf9eac03e4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"6050e478705ce11540629697de21ae2a91023f1724a42aec032f5ef16681851e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"b3b62e97f591d74f61fdd90734586cc9fcd489bd284dc8c5f0466ea20c851d1f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"30044516a5d45e27a95e3939762cf58d0e96c6e770df89c858374bbe59455b97","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"d7ba5ae6c994c4780d1972f6c1818edc815dddc31081b0c1b69a41c4a0c6c3a0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"ce27c2dc9a88c701e2715e2e658a4cc795696078fd61d78b741bc51025610894","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"c953828534cfaa6ee58fc55891cfff210120936af0cbf627cf5031ff02b66dac","bytes":5258}},"requestSha256":"d5bd75aed5ca74f1be74a75b72a6f866b4ce11f184805d5ecd07141b742c9fa9","metricsSha256":"a78dd8e71b3cac5fe5cfd03d84254a70e9343341609eb24565e514b99ec48772","continuationSha256":"476de210a92ceca312c7e17235f15f34f452af5a2ece3f5c36993735129c634f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"3499cba5c2f9330d9fb4fa31df05ef692e2fe2bc5fe30411af8606f7730d6a26","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"b8c2f65a4effc4e48d4c3299f6fb54dc0d750dfddab4f8badf5e25bd432cf721","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"9e8cd05f0eb65d7e0968ea431247cc4542956f4dd21835724861213dcce033a8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"90d9f01cda12261fd69ac45fea2131c3459c10a8ffc5a665c1222daf9eac03e4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"6050e478705ce11540629697de21ae2a91023f1724a42aec032f5ef16681851e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"b3b62e97f591d74f61fdd90734586cc9fcd489bd284dc8c5f0466ea20c851d1f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"30044516a5d45e27a95e3939762cf58d0e96c6e770df89c858374bbe59455b97","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"d7ba5ae6c994c4780d1972f6c1818edc815dddc31081b0c1b69a41c4a0c6c3a0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"ce27c2dc9a88c701e2715e2e658a4cc795696078fd61d78b741bc51025610894","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"44502f767cbc04f03acc7f2493296a342bed7fc6a737a8caae5f4486cb279539","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"03caa69fc2a2a3ad674bd7f4d5f32a722d08e5d5ae9a9be7e5f6ecedfd46db96","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4c0567ca492a2d7b60d1c8fbcb75abeab397537c2f8a26a15a5019da7c33da5e","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2482599","recipe_identity":"8d483290942b186734c93c48692b178961bca881fd92251c6f3e15e44bf4de55","loss":3.425501630079509,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"c3d7cf7ef9548cae6149c9d954e49081c36257be3d9e20c1a06d71e28c1df49f","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"be27aaec9c598e05391d3aa64e7616153a8a0d2579aa4cfacfa0f86f2179f847","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"c9dbec1d7fbed1609d78683c84c7e7791fe5cfa73e67a47b4046c6690390ed99","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"7df3d67f53330fd27f702c9b23ac7c3f40573d4f3d902232c84bbfa6b91f75a3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"8115fcbc2f400995801a03c967d7d30f10c753731f3108d3fcde360c2bc0860c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"52cbf8ae0eb9cc9374487fd8f9a6be7cb3cd88628f5643c0410b9393afff4a2f","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"9d7bb97462276cb94f1bf18c03867e5d3502f5d2acca9c441c0a61f1589f9e56","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"c703dc9884d66a59a53609498a0d47f3f9448567baa3637534e776487e47e9bf","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"a37189ded5b3245a5a617730adb7b79c1747b1d4913bd20c91ca90aac082e6ca","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"795c2ffd107188a4efb9df1c81912988bc5e3654b46a5c98ff015d567b3e3f4d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"236315fe77d12bbca5c6ca0b973a8261255cfdc82157404a7517b03611734be9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"f5f130a598b047fab53c94e1643b1dff54e0c5897608ddb8741ec943b6adec83","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"fc892a6bbaef19d66bec6e49d3baa2dff6d23a1065cc3805955c86894339b062","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"b484658a0984b382f8d6027e182766a7ca9c4a80218fbfc7acc2597944809460","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"026ee95ade59b25d214fc9dfc9b187e36ef2042ea734ffc85b60619cba85688a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"98ed3958f0affb2ce61b10808c34905ba0728cae13c1a2918a6f579b68d532d7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"e76e43583fefe44c8090081bf00bf4f3fa135ac7d0ae1703d3fb83341e64a7e2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"57c4413fd08e71777dbf47c3591dbe70ee9e32d32117336b7be132fbb0a011b2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"c6c98a3a9183e9c51dfe1ef3e2d400543cc69a08e86e5d5fc1b72b5221a891ec","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"faed6b3a46039c06ce6a69bfdf5f0a0a733a939dd37e273dbc17f7f82e4f1822","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"ebcd9fbb7a36576244370e23d8ddd78e68145966768664e7c01efedbf1caa391","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"36797eb8de0f6f3ff78304c9b124cc56b4f67d0ec155af95d6e69d8ddb5b7482","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"049c53631d9531a32c0a36d6160206d829d7af47426f06f9f59ca329444154f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"6be3da0657f93e52db58df4e2d089e3b24ac5c588e602ac4a04cb8fedf97c27f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"8fb6e061bfb2b550a71ba548b7662ff497c4549df86492109dc3042ae9bb5322","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"ea6001d6faaf7d7a831a676bcbbaadcaa9b81025798e1d1e1fc88a638b6bb88c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"5971765f65f4c7e6f52b35818ad0097af56eb21d47f2d4616c5404c3eb48ee3f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"528e1422712accf98808433fdd0166b5a95e564ecc7d6da05379c064199c1d71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"d35a4757b6f4f6abf17b0c94dc5868bfc2e85b41db449d7bb7fa1ae292b4f363","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"6b38af69c2f44ae30e8ddd340a925cf98d8eefc5fb824dd215d6cefd17ed2904","bytes":5343}},"requestSha256":"a22c18060629715a2530b80fe7a154b6fdfaceef6738c7c2f37f048667caf9f5","metricsSha256":"8e04fd75d8cfb100797964e3e749a509b153b0b8657bf8ef8bab90af573208b0","continuationSha256":"d8e155d39e3ece0b7491603746837267564c96b4bdc4a3716f17a702db2a4cd9","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"ebcd9fbb7a36576244370e23d8ddd78e68145966768664e7c01efedbf1caa391","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"36797eb8de0f6f3ff78304c9b124cc56b4f67d0ec155af95d6e69d8ddb5b7482","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"049c53631d9531a32c0a36d6160206d829d7af47426f06f9f59ca329444154f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"6be3da0657f93e52db58df4e2d089e3b24ac5c588e602ac4a04cb8fedf97c27f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"8fb6e061bfb2b550a71ba548b7662ff497c4549df86492109dc3042ae9bb5322","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"ea6001d6faaf7d7a831a676bcbbaadcaa9b81025798e1d1e1fc88a638b6bb88c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"5971765f65f4c7e6f52b35818ad0097af56eb21d47f2d4616c5404c3eb48ee3f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"528e1422712accf98808433fdd0166b5a95e564ecc7d6da05379c064199c1d71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"d35a4757b6f4f6abf17b0c94dc5868bfc2e85b41db449d7bb7fa1ae292b4f363","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"9416265d3a5e8bbcec4f79f508de41bc46122dca7b318c364c593ad628d7a237","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"944a04c9a3f189941e4794e5483996f108182266d8dbabde0450f8c758bdcf10","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8d483290942b186734c93c48692b178961bca881fd92251c6f3e15e44bf4de55","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2482600","recipe_identity":"e27ed699c1199f0e866a57bec7f05eb05c665d880748ff2d824569fa6645af04","loss":3.4301407902846344,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"35a12d15f48624ee2cb405ec7ce4f555d6b513de8ca746be7daf93cd501ca3f5","bytes":883},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"f0fae7c0754dadda0673164b20e32862d0e11a8bafccbe3226cc3b6862bd8e85","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"4b29728b8f914e395b3bbdb9156633c82079364cbbb52ca6144c0d677a34f1e3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"94f14be832885fa8ab587d31971b8394cabbba0759e508266851b79ee481f66a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"adcf7fded442f4b16fb5365fe8121011db9efeffa76729e97daf82def4cbce3e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"2e0a5af67084cb4791a122e026d9854835d3c89a16e2940f0bd327658288ade2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"02df77afac247a34d6127b020d88ea007a1a8cf8341263987bf37cf9aa96a62e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"7ff6917e014c09df1a60b4def72f3d804d0c657717221367e3b23d05e1d350dc","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"15d46b8bdf9a736854bd288010a2b4824e9cf27ecca3a9cf410f9f3711506bb9","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"ce67d1836d5bb26a1f0a73f091f1acdca93479d33fdc2c8e3e57036cb4d75ba4","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"3c0434c6d6163fe214b09771ffb2852fdb2218c1d18f1a0c441d72576f818bd1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"fc1f9cee24636715c1b2f5fa945202b260f0fd89369c6b58aa2cf45eabf34e7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"f405da431c5eeecf2d7069fcb01170698a3d74c20b5f97cb69a27ad15e7b1a17","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"0c9650b709b8428e6584ef704887707910cb556f71ad8808f52ee874452e1989","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"76ccdd9cd11627844acbd1f5abf2535ca6507c228e3b6deec5dbcdcced4999ea","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"5749bc2a170a2785571f31bfa5096a0f672fd4bf9cb73b166683692399e37b91","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"a1543867623d89a4fd8df640384811ebe6b4e86b6d44683f62f8998fbf8e1a67","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"a373855e2103ed11fee3547aabcd74ce2cc5a3ba86536a6f5a54d50338626410","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"864c556e3345995233656a6fc4d7385c317dc6abb0e7b9f67bacc9d3ff53a795","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"617ef33fa8b558c6a0c92966fff8fdfa5150661ad746becfb1106a93ed9b44ea","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"dbdf510e8d1e3c1bdda2d2bee1a1cdf63c589138718ffe75852d4265a21524aa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"cbacee768f95f6672e1ecddbb6eedca14d40b6cdc66e163207ebfcf0126a96da","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"b9c19d348f2982aca0542bf5d317677ae50e0826d121c55e2c60923d539335ad","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"32c44c3bfecd9272912460acbf8f160c46631f6face525f1f709aadcdf540106","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"99dabc2a1e7eb79b9b7ab05bbba379fb08a724fa51a685d227cd168ab59532bd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"a750d0dabf0cb17fe84d2339ebd05ea61cb98dc2cf152654d3802dbff3bd34cd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"432601e8ce9fb9084a4cef33babed2525e947d00d43d44fb10036f6d36732084","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"960c587412df79ea26020efcea6db8cb71a7651c87d3b87afef9b1bc8d2a7a1a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"80969b0147861c3658970d9a5f61ad014001f837ad5a71109d963da164919de2","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"c0e2786e21105d86821186ef3bebabccfd2826d913c66166cde3e8a7fc8eb9b0","bytes":5343}},"requestSha256":"86c166f0d8145a6518a0dbe63db713cadc582bc8945e5712491ba533ae4b7697","metricsSha256":"a2532aec89b5782428d93a01597127c9061620c78b52f654ea99cdfacaaa655a","continuationSha256":"2cf9ffdbdda4e74d02f940455dd160c68dca65cc571207ee0161f11199ea81a6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"dbdf510e8d1e3c1bdda2d2bee1a1cdf63c589138718ffe75852d4265a21524aa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"cbacee768f95f6672e1ecddbb6eedca14d40b6cdc66e163207ebfcf0126a96da","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"b9c19d348f2982aca0542bf5d317677ae50e0826d121c55e2c60923d539335ad","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"32c44c3bfecd9272912460acbf8f160c46631f6face525f1f709aadcdf540106","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"99dabc2a1e7eb79b9b7ab05bbba379fb08a724fa51a685d227cd168ab59532bd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"a750d0dabf0cb17fe84d2339ebd05ea61cb98dc2cf152654d3802dbff3bd34cd","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"432601e8ce9fb9084a4cef33babed2525e947d00d43d44fb10036f6d36732084","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"960c587412df79ea26020efcea6db8cb71a7651c87d3b87afef9b1bc8d2a7a1a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"80969b0147861c3658970d9a5f61ad014001f837ad5a71109d963da164919de2","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"dc5bf34c7cb491e3c6ef087849645a77e99790abfd42d5612dd8bcffeb361a94","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9530a1585cd21cfb097163755126210cf89e860866d192875e180c1e015b13f7","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e27ed699c1199f0e866a57bec7f05eb05c665d880748ff2d824569fa6645af04","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2482601","recipe_identity":"068ff7ed9340c26aa55ec2db14309c51afb30dac74875affd32bc26110133bc4","loss":3.3777699664737395,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"3fef65f5809e21948bf8cb6071b311dcecbde7d113207757981ab98687db654d","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"bee370819fc71e621643401772b133c87d790fa3f8f6cd5194d14ef6113ac912","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"97e9911ebeb940dc25be80717743e57f83b0aa9fbbb61290328bbd34bf383ae0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"98d1540ae548073706a1f9703aa79b0a99155c75d6ee50d1f8144dd6c4bc717a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"447e27f998c5c5b04111e8ec0539e2395a482c9a86df1a592633f84d750cf81f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"a37649b47755d45400b79b3b7d663f6373f99127c7994283deedd36aba5dd22c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"f9d44ea26ca0310a948ece9b3799ccdc8c06f0ced0812dd8ec746b4486ea1ac1","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"fe193efad58fb320777d4f1177649ced229d9a2e1533bc22d19abd734c04665c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"91200dd8b27bb9d7ba248566af1e5a5bf903a1e674d1bbb54f5ff7d34c295f48","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"74e75645ccdd9c310434f45c46ca71110b56e390c27129706bdad1e016fac324","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"f3bd58d1c7b811288f160356c5f2c447a148d5caf07967ccf7d3cf1f17f71601","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"7058ee955955cc38232782421355f938a61b7382de5ea48b9a487b4e610712cb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"db0baca5f56edd6d2a239e71845053b9b5258787c1f3b42645552ebb71e8441a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"5fad8cec5a4a14c7e94a6296e8ce4208ebc99a2ab1ef419a35a0ab059ecf6555","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"2125eb767af82733c4572c3a3cda8478438222bef312dac58eca5cc76a6f50d3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"49c8051c2239c3ab75dffa2f7986fbb0fafaa6c68c5271c6e8678c7f7d5dad7f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"22a732f47217215b46c4a89dfb4543901a102e3015fd89d77ec03d8f33049d54","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"4fb65c55b463e28f52288d06c239855400d06b9d17548160f5c5ce0f29670ab0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"ff189d31e41e67782c652c27bdff8c5e670926d2ed23426a51ea808c0d5a8f87","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"472ffb7726ca3cdfce20263e14a9628520d22dd12b2451ab35873ad9c825ddec","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"8529ce3c33b9c148b1a1675c6e357d3d57340659885fdf8d47f4e25a43de370b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"9aa77056e06d0f71f4b164e72400b9220d45232ad23cb763e3e3321924ae83cc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"8532b19baba781fed8926708c096fd57233e1018a847a3120bbf733e4c849400","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"d583d9658ca47df05466eb8a04a4e959bc2b762242caec0de4ab43d762a02aea","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"e266d0bb9b95a130c5139c2a90a04cfc9c34592d9438819d1f0ee5c2790be792","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"9cf119769c4aa79d5d880b99d0612b37bea235dda71bfc18dd1ca45e88794dbb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"ede226e60f857af1bcc21bce704e6eb99fdee0baf565e2ed476babd0927bc2d4","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"53feac1538429baf1cfc0296213859fa7b574aecb59d5da3eb928484dbdab5c0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"854b3f764d9f23ecd2098f8bff58b70a1c21bb2a1ae909494a1c0c61dbd9566c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"19190fd5ff45f3817a885102eb473d611e907b06df6c61f43622a6e6e3c16f4d","bytes":5343}},"requestSha256":"38376f1721989c35de035b21c8f4268027bcc5b77318c9350ab97c81e2a23bb7","metricsSha256":"e2b3671e9a701c2dd76671f9bc0d805704b3627d0b910c81240e6000c3aa6cea","continuationSha256":"c3a56154d3ce5156d74f0a6492b273b9bc30dc8c35787f8aace09c94dda6708b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"8529ce3c33b9c148b1a1675c6e357d3d57340659885fdf8d47f4e25a43de370b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"9aa77056e06d0f71f4b164e72400b9220d45232ad23cb763e3e3321924ae83cc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"8532b19baba781fed8926708c096fd57233e1018a847a3120bbf733e4c849400","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"d583d9658ca47df05466eb8a04a4e959bc2b762242caec0de4ab43d762a02aea","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"e266d0bb9b95a130c5139c2a90a04cfc9c34592d9438819d1f0ee5c2790be792","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"9cf119769c4aa79d5d880b99d0612b37bea235dda71bfc18dd1ca45e88794dbb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"ede226e60f857af1bcc21bce704e6eb99fdee0baf565e2ed476babd0927bc2d4","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"53feac1538429baf1cfc0296213859fa7b574aecb59d5da3eb928484dbdab5c0","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"854b3f764d9f23ecd2098f8bff58b70a1c21bb2a1ae909494a1c0c61dbd9566c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"9aef6d0c7d1d46b569a07eef81375e0b180d61a2b2a7b83dcea823a77962c3c5","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e642983b4093fd618803c2e6cc1d3c6de534ae2e0cc2cb09f1c1d30c06aa728f","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"068ff7ed9340c26aa55ec2db14309c51afb30dac74875affd32bc26110133bc4","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2482602","recipe_identity":"40f9eef31dd0f2deeae7139ade500078d79fdb61354a99e01e4f7d45a46280bd","loss":3.385437664644348,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"cfe60f3704e605e46255b70560677186a0f1d465e268c8d4db957f28f1512076","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"87448b63aa394cbad8ae89dc805f60d0ccc44477d010d3db910449c9ecee19d7","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"a0a7806eeb59ac72609dc293c3dee9806594fb66fbc5b8a437b09a0d13fb1a03","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"a5695312547949eb6c62927c2f223691461ae44e1d7856b73d9e5f332b7c0428","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"384378a12a963363ead23196b5680de93b259f55c5ffc66d6d2c436e3f407238","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"c64ea071d9477d0b1d9db726a1eb7c259cb05a7e00bfccd71ede74c7c5adc8dd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"00b8ec1b06fa4b6ef7f4d3b6f3bdd1610072be0515e74bff73a4f70f22eaaff7","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"b6fe38a493a9e6ed028708eb8f2f37e423098fd1ce3975018face9439938a8f9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"09b139a053dc8fbaacb36f9bd71c51c93ab2dde2bfb15a3dd085848e3236a458","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"0e2fd3e59338e5a03902b3b67ba3123fd590defa857208eb2ae5a0d2117315fe","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"aaffa9624d1252ea101f364af624f05a296da42a2d8baa99a5cccb0d5f6869a7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d38d74bc58ebec5a25b3d6fb1a7a7cccf8bca544bb959684820cff0182738e41","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"c689576dbd91576b6fcdea4b0a5bbf3e26c3dcaba9a0abbc304b920bab2c414e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"c5c08666e1d682babde7fcdda8af89eb2cd7cb5704c903299a28b0fccf3a4f72","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"e3493b45856e1844caf4aa795fb4fee64595ec0ba6ffd661237d45a673ce1611","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"c5248b56591ad3383f2b333dc75361e641701e1055a9455316442c7c4807a10c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"5a3d004aaaf771abc062293a993c1c9c51422aa601a68b323c122f5246beb4f7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"3ebb4a8941ff7c88bf86c646c9e37c559a0d82befe92c2eb8c7eb974cd231af3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"287890962691c0029e42efb9a78daa0cba27131223fde6fce9d9f6222cf57ead","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"af7e7b34aa2b994ea9133ddddf8a7c4e8b3c7e23d4e032760227e949c6abaee6","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"2ef7234451ad974b27b09ab7288fcde974093f0c1b2df4e8772ad935b671d9bf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e17abf529cc23e2d6f8daf829e5683503d156722fa4aef769a64da286372c183","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"0b6c7ace6148eb8b86a281164c3133a9f94bd9aada2c5d4507208d18d4286cb4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"51cff3f7edf683b6fc5eae1c26b0497b41a572f0e948827bdf47b49aeaa042d7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"43af1a4240dd01040fb9ac2b98f698063f178e617eda2a6e1e004a0e05d7c938","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"b9e059463d2a2657acbf4484d63ec75451343ed579315bb56297f379645d90a6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"de467b96f0786100c2302159719902c00ded203f2ebf94ab6e614dfd020d799c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"7e9be0c0cb8fae030266dcd3cf03d11a7b78b51a347647bc99e9eea1580f3f77","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"cc13c0e6f3e0c3abf838904962bd3790098d2cf9c998e58d5a2f691abf35aa31","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"7ed9fe857db0f6a49ccdb207f02144f5a4e21a0a0b3cda3b39ef37972e74cddb","bytes":5257}},"requestSha256":"239d81e270a8c30f97e42b105ca7f00cf37866a6567e8094ffe7dbb46fa9b9f0","metricsSha256":"df010bdecf78fbc75d227ac265874a6569acda89306be9c467ccb2238e5218fc","continuationSha256":"c27b8d1ff185403c9159c94eb40f0bfda64667305f68cee77c55a3aba5835484","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"2ef7234451ad974b27b09ab7288fcde974093f0c1b2df4e8772ad935b671d9bf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e17abf529cc23e2d6f8daf829e5683503d156722fa4aef769a64da286372c183","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"0b6c7ace6148eb8b86a281164c3133a9f94bd9aada2c5d4507208d18d4286cb4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"51cff3f7edf683b6fc5eae1c26b0497b41a572f0e948827bdf47b49aeaa042d7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"43af1a4240dd01040fb9ac2b98f698063f178e617eda2a6e1e004a0e05d7c938","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"b9e059463d2a2657acbf4484d63ec75451343ed579315bb56297f379645d90a6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"de467b96f0786100c2302159719902c00ded203f2ebf94ab6e614dfd020d799c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"7e9be0c0cb8fae030266dcd3cf03d11a7b78b51a347647bc99e9eea1580f3f77","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"cc13c0e6f3e0c3abf838904962bd3790098d2cf9c998e58d5a2f691abf35aa31","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"c9901e36b404d51689f23fe2264310fc9de0851ae67f19b6db6383db4fe89ecf","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1eb91a952f4cae71e2d946688da5580e324d33a06aca7b85e6cfcea5d424a357","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"40f9eef31dd0f2deeae7139ade500078d79fdb61354a99e01e4f7d45a46280bd","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2482603","recipe_identity":"0c3a60b25c9262f6a013dae39ff66ab9413b6fd0137f10057971688b95e7d2d5","loss":3.3730460376330753,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"669a7c2b3735af434a56143d66511f4f2bcdf57a0af3c1f5b637214928e66d84","bytes":880},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"360e6ad7c7cf749572b67505da2df6f07f69d85e8ec81ca210b8906a12d0722a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"c485dcfdbbd814de69f2560e58641114ae9bae8433152dfb625f4f3fa685ba84","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"227f9e51cebae9920cab737cf64fac903b334c9e9b8ec17dc227be73d3c2e48c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"450c55f7d78ef55d40bdd6650746f4d5b6028528da912c5c07b34e86bf7e4427","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"a5c7d371419f38e8040e72da25979b6952c89ec8eec8a5bd4ec993e3ff2a5fb1","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"8126137140469f149b03b42668214a365e93d55c0ee52d4cf15c2c84c3cd4eb3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"75566fe7a133a046b4dca56e685d552d65fc804dbfd41539229c76dc86d3c25c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"51782cd8cc706c7a7cb4c3842f34ea9591ff8dc31d06c55a203c2a97d59836e0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"2f28aa07c86349260ea52bf5f169d0453fcb5fc99fa636300649548beb4946ff","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"d7649cdcadd1060830a08545b8ec6a2e6f64697c4b1c1367f00432f27ae80fc6","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"1e5410e971e8bdad91763f32cbbac7ddf57eb2201215e26b9526a4b9035922fb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"34e6a29a92ee2c5b83f17a8862cba3731df0d0b86455bb1d5a79b7f89eca6b12","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"b8528527037fc0ae3cbd7265d585da05cb0333c2798a73cab56fbc26f9c08121","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"c2c2d7210c244414dd43f78a0dd54d37505467f46d9b8902ded49eed7eff7bb0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"6bfbe5ec245c7c297cd06d2d17c15ab9ab3773d21a0f0df95e1d911f7a4dc8f8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"0071fe9fa269d7e01b5176ed098938cbc0d43939cfa33c0a30f6d6fe252d94e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"df39f29228842ebf91f0997327ff72982156e0a1db05ee5b20dafe7334b56eb1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"6808d7959bd5819b08cc963d613c6a1316b54afdb388ac3d33fe49aeb8aa4946","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"bb1255b75d7870bf59e94e9dc50cfbc35c0d5b151f168d904fedd0b2a7eed723","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"9330d048862fcb13451a2e9d8810242475a70ba268c3ec78ec1d9d29ff7c6717","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"d0639562967ef175cbd7d7f31494585f6486f3b0f8617702d0e6f6e91c55a2de","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"c96a7418c37dd4799e79ec321e54216d919d14be691876ec0ed1bf85d0d209f2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"87b61c00191af8702c13a8617d457df72089890e0c18e86dd369f0a53b8c1aca","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"9507afe2be1f8f6091c1d5765619b5265874289a960e950056cba3e78218565d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"24dc6e6677931c8673a1fdcc7482a2a79c3a5372894dc92225d43b5f29d46220","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"73989b670feae74c16753378c3b20f822df72369836156fd4f339d1b49946f09","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"f63f8070f87e30d8c316cd4365395a4894790bf77ab964105c3393665a6b17be","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"65a125b3306d9807d6e04c6bec76eb806476c4ef41135c60eceb796a1680786d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"d1614266efb475246f1c618d2a3dc8744070c199fa9d836461f7dfd283b2b3ed","bytes":5261}},"requestSha256":"01cf875e70be69ffad94e925e07b44ac2827a6a93f54915ff85b890b3640ead5","metricsSha256":"1f306e0daeffc91f29d8505da352586d5716646e5a467f6756fc35fde517a226","continuationSha256":"c7ace52f3065d2f56de39673bb51e1b469fe00fdcbfe3b9ee0657a49175898d2","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"9330d048862fcb13451a2e9d8810242475a70ba268c3ec78ec1d9d29ff7c6717","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"d0639562967ef175cbd7d7f31494585f6486f3b0f8617702d0e6f6e91c55a2de","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"c96a7418c37dd4799e79ec321e54216d919d14be691876ec0ed1bf85d0d209f2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"87b61c00191af8702c13a8617d457df72089890e0c18e86dd369f0a53b8c1aca","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"9507afe2be1f8f6091c1d5765619b5265874289a960e950056cba3e78218565d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"24dc6e6677931c8673a1fdcc7482a2a79c3a5372894dc92225d43b5f29d46220","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"73989b670feae74c16753378c3b20f822df72369836156fd4f339d1b49946f09","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"f63f8070f87e30d8c316cd4365395a4894790bf77ab964105c3393665a6b17be","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"65a125b3306d9807d6e04c6bec76eb806476c4ef41135c60eceb796a1680786d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"409e139dbfbf2ab97293db5c4a71731ddf4efbeef27556f072dd4d045ac98b48","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ea2bab4899467f51e038fbb2f95d9b6be5a71355d311fe0ab7e0ba3a17f1418a","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0c3a60b25c9262f6a013dae39ff66ab9413b6fd0137f10057971688b95e7d2d5","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2482604","recipe_identity":"e4dc93c704e9170517a38b072bd7c723a99e8665b42c1e021f8b133fefe33ef9","loss":3.3875030628144307,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"0727930a8eb216c0573cf307538f8bf5a435b20bbaeb7340ceaffce4fce47211","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"807ef2fa7d88e30e60fd6dcf05860d9f76e427cfd7df6058dab31b56c3e1da65","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"20dcb6e1523ed9ca4d67dcd894aba10d284788e7857f815873d7d44931ab1d55","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"f6de707c646d25802baa7ef2a3083d0a3525f3a588b5695cf89dcd6ce608b4ce","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"51c15bd3e61d1a635e109e3d521b36a838e914c8cefef99a623e8721ec2287bf","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"6c780d517ab07a7410432fb94967fd32a8c128e52feb22ddf7de7dbe0b442c0b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"9f212fa41b1e976adbf5dc1a3589beceb2e7e4b461eda1070cece120c766c6a1","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"f1e60a316ab20a70034c06251def09eea24911b6715d61d5f482ba89a2752408","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"d2c7e3753ea012b9a79451b9ce71ff368ce3760914fe25a3bf17d81016b67475","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"b42d757929a8cf5878747398181b4171cc125aa434b40eafc079b2887daff0de","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"de48e41a9d9c736ec68e6396d01e601bc5dd2b55efde4853cbf09edc28a43cfc","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"8ab77603aebb6e35f3d63562687de24f8ecda661e6517ea2841c1ce524d3b9e1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"fa7d901f8b62d73d321d0d8a3ccd6ec09cc16df8c441af4eb61ebe28663928fb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"9d65f86e7c05548a235e38e3a986b2b12aeb6bcb5064706ec4bfcf84230ea229","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"f1c0b01695ac240f642cd3649c02f5e0a53b91560d1015bc71ab9e861e629364","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"a4f5aab1f7ebb40dd6def510d64aaf189e3bcd04f4708d839f78c5f2a6046b57","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"74c2cfb733a86877b2dbc2a03ce78900b13e0b44fb30c35e51ba7d6284085b29","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"f468f4dc728469d582321dcf7cd5a83debfc3b655ed911971cf5fd72a7dae4c5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"111e1a66bb476e9a63ad7ada579302ba1a9314b84769581900c9145952f0b409","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"2b84f104dd799c391f7524b9f7d079aefec91ff711882fc3154400d2b17fc9cd","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"cd7d47f25a0ae8fd339961218134f497b69cbfc4e261a4721860157098cc6462","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"adfe8778e18962676edfc433e0ba1f113e18eb8f5a04daae2be094e45a9a522d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"262b06cb67ca70657624937ce31daf1a13e51d251c5d9ecdbc47ceff8de55d61","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"3a2061c2c0721befc197af7a965ba1b99fbf9a8e04c715dc5ff08ae938d18334","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"36e04b1a1f174cac5386289c5577d09e9ee459293d5a2d6c1440051be6aa6760","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"9db492cac366a1bd58740d64f964da9edad447c6217bff32ac09b62c015f7e36","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"d592aca80ad4cd59652276674345bd00b3642d288e298fa132e735d51eda4750","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"c6d41d7b825b705f939c0028cd7e520f007993cc81151c3fad01d59e5b1455c7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"f1ef767f4f15e1e42e2a35c6e0a89fca8031bb8972573ccd300b7da2d831a53a","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"56b4322ae68edd45d0b8000421ffb74086dad50bd47136b0a1f109d502de283b","bytes":5314}},"requestSha256":"ed0bf364d82b4af9404f092b867655f38607a043ea178a91cea8ed773d0078d8","metricsSha256":"7e434fc0df6e4fa103a92246d46f36f479855abea691616913c30440eec057a7","continuationSha256":"bc960d7f1993e0b8bf07554f3b5d1bf0a1aec3c3dd50ac8a78f6e0fdec93056f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"cd7d47f25a0ae8fd339961218134f497b69cbfc4e261a4721860157098cc6462","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"adfe8778e18962676edfc433e0ba1f113e18eb8f5a04daae2be094e45a9a522d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"262b06cb67ca70657624937ce31daf1a13e51d251c5d9ecdbc47ceff8de55d61","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"3a2061c2c0721befc197af7a965ba1b99fbf9a8e04c715dc5ff08ae938d18334","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"36e04b1a1f174cac5386289c5577d09e9ee459293d5a2d6c1440051be6aa6760","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"9db492cac366a1bd58740d64f964da9edad447c6217bff32ac09b62c015f7e36","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"d592aca80ad4cd59652276674345bd00b3642d288e298fa132e735d51eda4750","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"c6d41d7b825b705f939c0028cd7e520f007993cc81151c3fad01d59e5b1455c7","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"f1ef767f4f15e1e42e2a35c6e0a89fca8031bb8972573ccd300b7da2d831a53a","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"a4b725d85bb5ed02b564a41cb4c10770a479e2cfc9e571778c0308d9f8533b3f","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6a3e866907f811331edbdcfa10bd78447d6b5ec53457ad94cb50b22945353943","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e4dc93c704e9170517a38b072bd7c723a99e8665b42c1e021f8b133fefe33ef9","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2482605","recipe_identity":"afb0cd3807eb251d04fb89becc78a6b8ba8d2d8126f38d101f47f7d069bfadfe","loss":3.386587167274686,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"65358cd7fb97f728949440da4385c9b600a1d4290efc2add9d8e93adc81e3d62","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"6ea2c37d72784a9f789b0f38b27f8cc95aa33dc0e2b3bd4cc7bb1f404f7d38ac","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"e427236a83d72a4329b7fa626ffa5b4c08ecede4fa6814cb56113bf05b5ea238","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"152a3c9f367165252247effa31dedd61714e53d33a8f45be9e60dbf5bb91a927","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"2ec5172c8dd44951db4544d083429ff7bf034c2788f159d56d8cb6a9b0a641b5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"e1712a23555a44434dda818e1134d9c06583abcdaec42cfbcb8ecf50beca72e2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"949e12828d00f9b8afe7cfc95f3769dda27347a9cebfac28b50a501a8e7ae8cc","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"e6e3bdb78ac44e848de2cf1e5b63fdb9fab59bc52f89819c509c5506a6ad5d73","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"2426a2ea34af468733c24f0ae7f325beea22cddfa3b44fcb78b9c02cb00636df","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"bd9a5fa9ad0069e75afa3d0a85d8acdbc3bc5b6c9d0fd157fd57e8f5a63ef29f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"accbd6d759b3479bd4f25608433e572ad39f95ee7e5540001071d1d19a9b8e82","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"b80452cc4c17b9f27a15966c35f126bab964743ee7c8736e2eaea2a2dc73e275","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"62d3ae8e3bf1f82a9f39b0cb27427962d4c176b87c4018a26e0a64457a905e40","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"0a4b4d0477c3cf116b36b4a4edb11efdd5d8c4fcbe2196b29c1bb82e4ddc7041","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"156deca5ddbfc73cbbea7bbe5c68d3f4aba377bb39040e5e8e6737420f82ae0f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"4f12936bb4cb4fb8a78cae25e6d0101b309956a080017f003d6192c7e2fe84ae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"baf45ac573c1222af790b7684a7cfc8e917fd02db6f4a740af5f58e30f71dc7f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"a586abbf31fe8464a7eac54e4fff52caf10cc736f3bd023c03db962c433abb89","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"11fe18849797ddb069cb438d861dce92211b1babc3ab3f751948e88bfb27130e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"4d0f17500ac68db6bccfca057040a00657ccc07e02f11d0e6f894af2b61687f5","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"bd185dd7c1680d37c70c26ea0ef8a3e519f6a79be88a7024adff7f14fe0cd214","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"05d82496ca0490d86319930b31146e45d2ac1abf4706cf00435d72d75cc4eaaf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"60f02c95429605848d162efa6cce4a41f9dd83deca21e5a2942dfe80891543fe","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"8326145ae6fbc8b786cfbf63de3ecb007a7c0069293a5287ec0c800566cb16fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"c67a3f3eaa356c11c530d6caf96f67ff7c909da65ab4bc99cfd3f57ccf6ac64f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"cda8ce9e3cb3a5cf8ff95db43f5486d5acce747f00fce2a04cc1a3d571829748","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"41fd124d127c34ffff6cf1232c633369c91cfa1ab2a9ae53049b935cd497dc16","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"05234741fe118d0683a58de45adcb5b585ba5f3c4f6369c138ba743f7fe45bf8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"add487440fa278d967d5ca1b484f2e362c7ddfa49edc31532c6dce163bc421d0","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"22dcc54644a882a14132f4818690f2d9891d88fa0078b0aa060098cbb178086f","bytes":5338}},"requestSha256":"1fd225e215ea4a77bd4fafecd50251d00a754cd750a885cca0115dcc8bdfd6c9","metricsSha256":"4b3323bfa2bb2164f640e60661451c5b93bc662320517fdb2ef4c4a04335aa1c","continuationSha256":"39b88e63ec6296df5ddd46f1c43e2ff2dcbafed58f42cf0233295f07c20e1fcf","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"bd185dd7c1680d37c70c26ea0ef8a3e519f6a79be88a7024adff7f14fe0cd214","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"05d82496ca0490d86319930b31146e45d2ac1abf4706cf00435d72d75cc4eaaf","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"60f02c95429605848d162efa6cce4a41f9dd83deca21e5a2942dfe80891543fe","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"8326145ae6fbc8b786cfbf63de3ecb007a7c0069293a5287ec0c800566cb16fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"c67a3f3eaa356c11c530d6caf96f67ff7c909da65ab4bc99cfd3f57ccf6ac64f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"cda8ce9e3cb3a5cf8ff95db43f5486d5acce747f00fce2a04cc1a3d571829748","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"41fd124d127c34ffff6cf1232c633369c91cfa1ab2a9ae53049b935cd497dc16","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"05234741fe118d0683a58de45adcb5b585ba5f3c4f6369c138ba743f7fe45bf8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"add487440fa278d967d5ca1b484f2e362c7ddfa49edc31532c6dce163bc421d0","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"53c48156f6f2e1983704be7b6475719e640243ab8787e5dab792f475bbdc1120","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"658c9cc7a95570e5cf725494839ae854f47c13ee2328532a9f591d2f8cc079d9","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"afb0cd3807eb251d04fb89becc78a6b8ba8d2d8126f38d101f47f7d069bfadfe","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2482606","recipe_identity":"2bcb0ec3776202a9a5f3de54d3107f13cc0f0bf929daf7a882cd6c97f35c1b65","loss":3.386050278247549,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"74901e34b46a63919888d55c2e3aab6331085721d5a076164c88b8ba46de3862","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"37b3deee2aebc8c169e922ef971a5ab96e38aa280aa3283481ae479a8255ba18","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"bf6ac202b36ee6a3b2f60ffd08ce016fa44c271bd1e89033e3a5e62bf8949f6b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"547adf76b896b29d03fe546d462f5e453e1bac8cc88c6c847af70bf57d70c449","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"2e59a20cf358635c3528b2f56c8438542eb841f2392bed2634a7f332009bf480","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"e78124dfe304e868b55f40af5a30d33cdb02e1d3f3e992a5009d73a7607899ee","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"8723e706df9f91913248ed88d4ec7456fe28e19c9b1d2ca7150441556ac62764","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"88f631b49909144617f1f03fb01e9d5c9b4fa1924e853fca2d339fa933528272","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"6a0de9f8050856389fd68d526a731f0ebd83566aedc6cb3273736439d023e864","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"9b51507ed241b0ef5f75e7be6d9013eb6604ad602be6c92ee4c48107c0fb2099","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"d840b9fd755c151104a5fb5e12156b3750ff59416b73ad02629fa9f3098ed81c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"d38cbef96511d8c6f0138d4b854ba393584bca24d34dc410fdf7fff8345a5afa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"e7411307926145e397e2b976b9c1d2c87339aebb0e1bfedda5af9059843c20cb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"fcee2a22cf9a0cde75520ba3b5c4e6ac70ef16202194763564545b3933c1755f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"2aba7a8d529e574c5cfbe8dae8539400cddc1afb8d4ab742ac7f2ab77402954d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"49aad23615aac1524f8ec687bdd3ef9b75f5186f82f380566bb3d844d67d21a1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"0d046dd8954c923816638bfdf0a189d2e1dfd3ced264bcedc83a95b835ebde5b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"565a1e1cba32aff2734d93ab527861722bc3ca37dde802e45c124aae4e516625","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"4474b100d90b904b845bda99f1b83b22380a791f79c0ac0e40b37d9165e71b90","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"481c91f9553a45dc4406ac187b3f89c3097e8475397a40469055b0def1302a17","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"4d1f8b71f39c233f48c42083d927e1acb129dcabea5df96b151b2a8c224f8041","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"b2753df54eeb3de9b05c2598f326cd2e0c0e9213322064d137c86aac03d6b52b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"a6d3a8e70cfe70cf67ee64e76a22a6468346d099dc956c17742874655fc77ce0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"bf59102b2467031b9c99f29b89be7dd47eadbb06c5c828bad63a875871a44a26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"0ae00b831aa05057e433b94a162d56824b03974adac24e5bb27c3b0aef8e745e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"6ce200b2cf3fc82711db480f79a441b10d21d98f9c98abcbfea6c957a1d1ec76","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"81a5b4977397fbbc9c8a6915c415fbd3029eb20333e68d1cd2861b6f4605b340","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"7d37e9c5acf69e4ec761ac3e118ca8362d31db589beb8fd7c2b18275ac1dfafd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"af8b88eb46d425bc002e221c5c3d1ac9b5a8d25640d6b682f29cf8bb7f2f9df6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"5867482f276cde09a529f875c97d6c46857de850eaf7b94758cedbfe404fcb16","bytes":5341}},"requestSha256":"f52774d971c876304974de12fcf1346eb85d4e1b9783a2ec082c0e734e125a60","metricsSha256":"d342df4baede42ed53326d20745c9109099bbe59d9a1ab7ff11910662e2decab","continuationSha256":"2b5f707bc48361f473160c5f3630f5ce24f15c87ce89bdff8d565419fe9f7259","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"4d1f8b71f39c233f48c42083d927e1acb129dcabea5df96b151b2a8c224f8041","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"b2753df54eeb3de9b05c2598f326cd2e0c0e9213322064d137c86aac03d6b52b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"a6d3a8e70cfe70cf67ee64e76a22a6468346d099dc956c17742874655fc77ce0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"bf59102b2467031b9c99f29b89be7dd47eadbb06c5c828bad63a875871a44a26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"0ae00b831aa05057e433b94a162d56824b03974adac24e5bb27c3b0aef8e745e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"6ce200b2cf3fc82711db480f79a441b10d21d98f9c98abcbfea6c957a1d1ec76","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"81a5b4977397fbbc9c8a6915c415fbd3029eb20333e68d1cd2861b6f4605b340","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"7d37e9c5acf69e4ec761ac3e118ca8362d31db589beb8fd7c2b18275ac1dfafd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"af8b88eb46d425bc002e221c5c3d1ac9b5a8d25640d6b682f29cf8bb7f2f9df6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"b03c531c004491e87ac3c83cf8b6d9f31fd29f279bc27338cc2b75312942cc2a","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3b645335dc09d815f1443d8d170c2a61a5eb999899713f8f2d6bf32aa047a211","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2bcb0ec3776202a9a5f3de54d3107f13cc0f0bf929daf7a882cd6c97f35c1b65","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2482607","recipe_identity":"d3fb75594397f749f86faaac9d8c6a9441c757aaa1299c24c52f625133a37ef9","loss":3.4565324680088088,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"b6053672aa084ca1c1f41211a4ca2e14150cbde69c0783cc0f31bcf381911eea","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"e69d0b3407e4e77362143e5c7e535b2cb27a96e7be246f5ce4db9160da8a9101","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"9bcbecd4337f1c51e1338fa488db98e8b11f30d03bb299baffa681432e1d1d80","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"fa72c392224ce20fd6e24efb783a654fcd6399d9d16aef74257baa9c680a486f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"370387e109ca90138aa8edb40d006afd7f0607fe55a3566fee5420d0f68aaa62","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"f935926b625049b9a54824cdcf4ea4820886370880e7a5fcb040ed786c8bf414","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"e062bb2cb00d55013fbdd128961023638458091a3ec2580037fb3a9fd6c4f175","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"3baee4c5894aa2df7d32d7cdc08066144aee2f9b37016e2903b53952ca956df9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"de54d23ae0cd8fdc692c544c5e8b23cc2006fa160693dc3abfbcf1bc22ec962f","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"47626d836aefc3d7e9dd1f87eeea072a50c85fcb0fa2a9736b849c130cff16b0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"68d0712c024cb94eb3ad87edbcedc22b8d294713645ba30b9907a81f6591c468","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"64e1237c5773790ac198606b544035c565733565b518f5cf24c9cae6f4e50bf5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"09c0aced6da6bdcb44ce250a3993339fa70a8b5331a252970924a53df65d0534","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"e4bd8f5a82bc647ba954536387a71fca1957cbebf52d075082721ef78dee59e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"5ffe3983e91758b7f2a0a08da92ba79589c83eb4a6ddf65c63cc456d828616e8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"ca3b9716d3023776ff41997e5cf0b6f3d333f39343841c558ede83be739f30c4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"2b2151820f76164182dde5e90fcdaf91aa9ddff10bf183465b3008fc57c194f6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"4fe18744afd7570c5a016a9e088af0ca2f8e75f881f349c1416afbdf2f26395f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"0a973f2e93479f28946d2ac77657b8d189a46dc9263da75c3b39b4f260754322","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"686f6040abee4bd33b5e038be5c39d71ec2e139016e55f7992e81825c1c5377d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"f29d6e3681247bbbea1e3c61dfaeb1e10a560ef05b0cc4299c7fb50291f463ce","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"db3e31716679e92fe4a27c3c9b7e6e6dffa268d5b1c57db80fd149fbecdc4a6a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"d088bcfb5e740f2f2da7388fcc9e07ced11534a6e4cb6e370934cea33f6bacdf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"055222ace1a74f5c18fe1401a5a5e41c76914c32f5fe68e740c3c8d2cbbda5cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"fd69bcc74fe692ead1e30d46aee4d50067d56e63c567211f27b4e5efdab248f5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"4ba4e47d8bf4fd5bee34fca2f87dbdb441b21e74587e34cca0a7f9fd60a820d8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"88839671ce2fa09d5d9f37b26974a02161937fdd739e14637856b60378d57639","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"ec32dbd1374af11a9de993e1ccba73f4b8ef6a95f9bf7692ef56fea2ee37174e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"544a21a7eb68b1a9c40b96dae09515bbb007905363be73dfac04bdd833d53749","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"bfccfbf5f8332b59b51cb2d1f8aade3dd921d8a8f0c6dc327c8d29b484192964","bytes":5252}},"requestSha256":"8cd690a3106bf79ae3bcda1a3fd112ac23983693b2360e7ed28ead08094e0968","metricsSha256":"74a0688692ac114cc81a3afc0e5e5f009afee280c54ad5594b1e6d4a571485dd","continuationSha256":"feae235d4b2494e5ef4c5181d5de534d4f48c506bfd1df7127ba8b8f8a19928b","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"f29d6e3681247bbbea1e3c61dfaeb1e10a560ef05b0cc4299c7fb50291f463ce","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"db3e31716679e92fe4a27c3c9b7e6e6dffa268d5b1c57db80fd149fbecdc4a6a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"d088bcfb5e740f2f2da7388fcc9e07ced11534a6e4cb6e370934cea33f6bacdf","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"055222ace1a74f5c18fe1401a5a5e41c76914c32f5fe68e740c3c8d2cbbda5cf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"fd69bcc74fe692ead1e30d46aee4d50067d56e63c567211f27b4e5efdab248f5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"4ba4e47d8bf4fd5bee34fca2f87dbdb441b21e74587e34cca0a7f9fd60a820d8","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"88839671ce2fa09d5d9f37b26974a02161937fdd739e14637856b60378d57639","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"ec32dbd1374af11a9de993e1ccba73f4b8ef6a95f9bf7692ef56fea2ee37174e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"544a21a7eb68b1a9c40b96dae09515bbb007905363be73dfac04bdd833d53749","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"75819607848a7a15228fc6a70e8a7521b9bc3aac9fff794b85b92a66c47f0cc6","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"aa4302c20a66f16330f92efe19f6fd6bb037b2e17f79e2ad53ced1b0932ff01b","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d3fb75594397f749f86faaac9d8c6a9441c757aaa1299c24c52f625133a37ef9","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2482608","recipe_identity":"dedf1d797344e95f44f5cd1d265acb62189451c7fd4a6f1085cb88f67a8c5bd5","loss":3.4592033680350687,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"f1684ceeb0bf1318314d01cb5b791c51b191dee2e84d85a3b5922ceeed40a518","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"fbe2141b84b4a7b3fa9526b95181bff1b4b93812befa0fd349e406f24f970239","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"d74ed8a565ed5178f31778cc076f897753d0ef6bcc89848c9c287da0bdbf2b68","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"36e406228dbea523cf023988716a1b91aa89febf4b41a29cb3f757fb11ad9e55","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"6476c5a371b6a9d1a8d6a1833fb3d254da939ef1d309eade16a2935cbadf379d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"ad0f284852f48edd7756d4958e7cf6b90dcf307f564405374e6f2ffe05380394","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"9df9261a2d3f343b1000f4b83bc545a3e3c2e4f3bad2c5e28ce8cffe34697bbe","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"e67c466404c202ae08bc652c0fbc3e8923775cee4f88d84cd4d2020f725b5d8a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"2333b9b5e12708976f972d24da5d54accbc7797dfff60bd64f6d5b1184bf5836","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"6bd5e75c95f4291fb27999e9c5d648a3495635d831b63cc95db4ee24c6623c25","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"02f7a9ab99e7a5ad29e1a07078f3ce41901a5db94db8743a55271fac601c4840","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"ea4b7a5231e24146c694622d8d05800d43e9572fdbcce0f7b77667e0aceb7707","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"2d9004f0ee8203f6e82fe4cb957714d5a2613c866ddd8ab95c0ba3a487e05bc7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"b08c68e8854001d82009487af7fcc222c2bec1d7d1be586751ae613f313e8b5b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"4fc078cb87d646181b86f297d336cf30c38d0f9c3a166773090dc242d83b93e1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"fc80dabb1782e568a9fa86d802d00262d08e2fe75bce9fb40a1ea40f0c716ace","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"8d14de57e40906f0a18d80b3373db847f7a1ceb2009f49eb64ed6217b6b7a635","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"624777078a7a3b2b5a0c883984c71dc6e544d073fa5cf5f5ed76c4692a1ed3a0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"88e27dd1769b117b8a4dfd645405c6781dfa0e20e6193adce741eba07f1471fd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"a2a2ec8bd1400a99965482aad22023a99adad8e8f75b207c114eb6d99e555500","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"97a2636ed288b3ff89ee278f531d8b9f474459eea7a05ea813dc1a70e909d616","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e8437defc7f3a6edf1ecbebb1ccd4577fb655f94ab587b4671b5999f7e09ed2d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"7056a2a55a5e7319cca26b93c64ba1959b4baf96685b329b8ac87c4c521785dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"efe4432ce0f0b7ee07dee4ef73f8e22afb98b6706e4c4adcbfbdce22bd61a815","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"fed7a39fa6a77ca743da8e87983e4817842fb0d063bca69d5c70e13697cd801c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"dd2be832e8a3f0227bd52d3790764369cf8ce8261b74369e62a730005130c3b0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"98842232e4429ab708457aa646f3cbc8aa46ef170cda6fcd7119a0916dadd23d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"0c792d5674348ad3209c2316f1d8d6c1b5c95a196d4712dd2d3be03c0c66da6a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"6dfac06ceb2f274e37a37428a99cfb1dcbfdc072c0bb5af1d2e577702b52f3fe","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"beedaf8f8db0fe189d99a3eb61db528473dd8b5554fb18ad8b01495b25cb5c18","bytes":5255}},"requestSha256":"a1d54cdd586058ee0b93dc59f33a6c2cc71ae0669e9182ed294e8be32e701769","metricsSha256":"d1b9f02ba6bd582d0f69ba5240f4abff2ba8a29a42edcc1086b0eedae86adfe9","continuationSha256":"fa71e92329e49be60f1fb86b949151fd079b23afcd2211d37616409b4263e842","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"97a2636ed288b3ff89ee278f531d8b9f474459eea7a05ea813dc1a70e909d616","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e8437defc7f3a6edf1ecbebb1ccd4577fb655f94ab587b4671b5999f7e09ed2d","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"7056a2a55a5e7319cca26b93c64ba1959b4baf96685b329b8ac87c4c521785dc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"efe4432ce0f0b7ee07dee4ef73f8e22afb98b6706e4c4adcbfbdce22bd61a815","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"fed7a39fa6a77ca743da8e87983e4817842fb0d063bca69d5c70e13697cd801c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"dd2be832e8a3f0227bd52d3790764369cf8ce8261b74369e62a730005130c3b0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"98842232e4429ab708457aa646f3cbc8aa46ef170cda6fcd7119a0916dadd23d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"0c792d5674348ad3209c2316f1d8d6c1b5c95a196d4712dd2d3be03c0c66da6a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"6dfac06ceb2f274e37a37428a99cfb1dcbfdc072c0bb5af1d2e577702b52f3fe","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"d51287e92a07c6a30ec8b65c7183e97b173aaac6a9ff3e77a6bbc79cf2e98907","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5e11520544593a2addcb4241d09a18ac6602a2bf5d87ae824ed9a8c31ae58c0c","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dedf1d797344e95f44f5cd1d265acb62189451c7fd4a6f1085cb88f67a8c5bd5","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2482609","recipe_identity":"4309c1d25bb5875d3c3fb05cf12be9850b6400b3bbcd07ad62d6916d04312699","loss":3.460333828039905,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"edfae55804fcb1fd67125c3a5ff00d41c7cb67ea42cc758ee09a6b2a4237b892","bytes":884},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"991d0b0f9bdb71283c05a88fe8d6c277296f925951ec924d421c1f97f5be73d0","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"d4c26ec0249074c996d7ee2898292d0a601efcf8e6fafd6ad58abf43df392a00","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"35a973dbee1b587055cbc715bb5a743a6406026d3144cdc7813670171a0a4ec3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"f84ce13df92362917613c1f31424d620d4d1f8fad8ef3285be879db09bf9600d","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"36ad1a256178d080e9f0c48c681d98c2f0f844c5941b37df2e98363efabefac8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"e417ad89fea6dc1ea9bd7f4ff01a196b26afe52644fc194d4df60613b8118e9d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"50158cf6e970d697f64dac3469c49afd5e039d0f8abe05d330260dc7ad457163","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"81074d7b3393556d95a5725ebac92441c4d8e42976f862ff9733382cdee4ea8e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"13f60873ad38af2f7eb13c5a011c20f382f8faf6b55c205b47998d276f46e9b7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"11edf0c0f3e5d8ae5a0b21aaed1b7be8b91af26b64c6383d6c1ef4067a20a586","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"3c1241fac308d4d5ebf5a3e070d808f55909437dcafc36119bce5787f6adf22b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"f79b1a4db7fd5e4c98b23c554e6e12cf500e3c4b5b32a9e613624bc8a2e21bd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"234937e4e816599d1ef9d8be977547dbae4d9d28cc27b04cc616f5d51da6b8f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"7039d85e3433834341cf52c0627d405b879e9c15a8c65b10f3af4ebe53d480d6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"315ac57ff80d87927e1bb591770a3c1118fcbafe83b6cbbec4e98055fd948e2b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"e7eb686f47106193bb53103b316a22af0b024ec53bc31e2ebbe86deb2afc28d4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"74b29d941f72845865413751324b8c1b5233620af246f22a350d55797d5ce8d1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"aa93d77dc8693ce9db38b754d0b21bcd8585fa2b89a430d1d1ba51cd19075a97","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"e7a6cc8af5e889083c59088c1f3ccdc2330960d0c7e02c9e631d564a1f9f3e30","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"a82b6c589d96dcb477c3900f42537fad4ecc6d373c6f2f53d89a5cccef1cbf76","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"c0c3228e7750b9a644c22e0546814b251aa6cc05df19fd74b02759f255908f25","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"5c8ccb80eee104bf0057728a16a26a09e18f143d88276ca51935326640889eb1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"ee240c5cdabfa2982d7d86491adbfe53a128444d3610882a0de0d8af6f652aa0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"8c1ef850ca0c49d59a5da969c467fd609075ea5bdc83961e339cf605504e9917","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"933ff467b2718dae7298fb58e2e0eeddd0de07f1634bf16bd726917eda03cffc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"a8b3c1b6bf6b2242b5d7168862622f6b312680e1a669f536a5e584ff78408901","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"af09672c7db4e9b298907bdc707a417c883353c6cfdddc29e7a6b57e25a5cd83","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"5f2f4f12c5842d27de601d0997a4a10b0a10392a6b503dd51d69ce4d66bb8b86","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"dd2885282a27738b7f22b5092a1a711929df820c35f459b1bc993196f2c3fd28","bytes":5258}},"requestSha256":"77c9b4acd2a0150dac2eaf0ec2751b07d71541ffe6ea7b0ea5434a3966dd4656","metricsSha256":"a13948354943a9368c454c34c4453fccfa5fab57585c4eff6677743307bf87fa","continuationSha256":"24079f450a63e493e37fb855e268039021cf5a814e94cdc52fc5dc69297d353f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"a82b6c589d96dcb477c3900f42537fad4ecc6d373c6f2f53d89a5cccef1cbf76","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"c0c3228e7750b9a644c22e0546814b251aa6cc05df19fd74b02759f255908f25","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"5c8ccb80eee104bf0057728a16a26a09e18f143d88276ca51935326640889eb1","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"ee240c5cdabfa2982d7d86491adbfe53a128444d3610882a0de0d8af6f652aa0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"8c1ef850ca0c49d59a5da969c467fd609075ea5bdc83961e339cf605504e9917","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"933ff467b2718dae7298fb58e2e0eeddd0de07f1634bf16bd726917eda03cffc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"a8b3c1b6bf6b2242b5d7168862622f6b312680e1a669f536a5e584ff78408901","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"af09672c7db4e9b298907bdc707a417c883353c6cfdddc29e7a6b57e25a5cd83","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"5f2f4f12c5842d27de601d0997a4a10b0a10392a6b503dd51d69ce4d66bb8b86","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"3a9d22e1049f69d5fd1ae801b62ca57b0052793d5f99648f962328fb47fed804","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9bf3b2e4247426c07cf9670d2da942ed6a66266f2e0b0699b70ed4c2e93e7c52","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4309c1d25bb5875d3c3fb05cf12be9850b6400b3bbcd07ad62d6916d04312699","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2482610","recipe_identity":"16144d3a7f152c2daaa9b461d29985e1805916d5b3fb1ac8dfcf0c5605c6c439","loss":3.466457894755254,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"c5ca05a0a9c6be724a040b319b2da8f3b3baddcae61dd40789726311d0662f41","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"9b26ac2787832943da8dfb25b7802997e276612b78e0ca6da134e7e2bdd77fc2","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"1e9dc7dbbda7ddb6b45fe83c4e0d6ff3c44df0da8261447d9c5ebbdc8b71ce11","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"8f1c34917993283078d6c12c54038bd2dc96508e67610a0852bb874013d23c1e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"6bb50baa79556ca4fe91443f804fcf961beea63e8e7f2f5578e333898bfcf6c3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"7b5bb545dadffde2492749e8f501843326845f40e7414c2eee46eec763e19ac0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"87fb297ac4159948298bc9b8a59213354bfb48a84d0defbfec997128b6f89d98","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"d889780216248518f767bf27484c04bf78b7a3a84821334345e1ef47c5798414","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"f826e709e0cbbf387ef884d833e729a968268df7ba0ce7b4c40f29e3e103d16b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"7ec6a442e235b867a0ebb6c1043586709a0186be488ef48584aaf0c1583f2b62","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"ccc7b55eb73ae861251a04f12ac9d3f6b72f66f53d57e99d5f6a369ec7680696","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"f3bd4780c16339fa56bf6163c7db76e8a1847c2af220c2c7c3148037dd5c29ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"2beeb10b84bb1f240de4627fc5ecbc795281145e1bac528f81660f47cfb4763a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"8e8207e01c54eb158eab1baea6d593c94b634912a8e0941f775fb7702e573719","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"2cbf6724258c3aa889659f14b3101a0e7497fb8a20e17811e97b409acd6d16b9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"29af7b47c136503ada4797b288c4dc07b6925032e73d276423c2f8d103adc849","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"c363ac2e8a9c2015fcbfa6a38b6811f425d8b6a47a774b79b335d9f55aea98f4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"9ac07d27a5643c0bbfcae1ea02b0dab0545b44a2f29760710b3827098046189d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"fa3f985d321c144dd22d9673ac957a5bc8c2f432049795a1d6528532f8580dfa","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"3884b148379b3d7450d9b0189c56d933942919f9f3c4db0e239e98c20f26d0fb","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"ca1423e760df68db29f94743b8aceaf9b8ba2538e054db6a35f53b32a81b1931","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"55fbbf3aaf51b7ed22ce8756db85d280093d3d9ff674019d7a3daab7bb3b8dd1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"a77201c61d2cb64402f76a3182013d6e64eeb64dc9040b0f7bf0bfbf7c0ebcf5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"9833e0e540a19075b4e5e991c6ad3dd1a3699007568e6840e6ef15dfe19ca04f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"856bbd62663c9ca8aa56bfe83938bb1f7522afa4192641fbd26cba9426071123","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"49d814536cd9fe5573c3e923e3152450cfb88c86403fa06938c427658b52413c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"604a165e1c9f2a9ad9d2459ac361311c73842cfd2b90b7f2f5666c691249ebb4","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"f5dc6095e432aa29c47e82603cb9564458465c53d674aaccd22ccaec43a07e71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"9e927bdcb6eb5957fcbaccdb025e77f32fbe0b9aa69f0a67324e1cbb752ddc47","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"7d26d640a3f3ac3b2e740c1f7fe09ebf4d9654a8294dcf9074495f2dc7b8a26d","bytes":5338}},"requestSha256":"c0cc6d8fbc14992137e2e4f83741fbaa28d3ded9bb43f45c27d04e6a88719b3f","metricsSha256":"820d5eb1c6bf7ace6011d3135d4261e8a97f1c8e13718104dd746316034dfaec","continuationSha256":"a66950f4bb5979a56b2265f03233f6f1103133af70a22c57ff993092080a6012","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"ca1423e760df68db29f94743b8aceaf9b8ba2538e054db6a35f53b32a81b1931","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"55fbbf3aaf51b7ed22ce8756db85d280093d3d9ff674019d7a3daab7bb3b8dd1","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"a77201c61d2cb64402f76a3182013d6e64eeb64dc9040b0f7bf0bfbf7c0ebcf5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"9833e0e540a19075b4e5e991c6ad3dd1a3699007568e6840e6ef15dfe19ca04f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"856bbd62663c9ca8aa56bfe83938bb1f7522afa4192641fbd26cba9426071123","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"49d814536cd9fe5573c3e923e3152450cfb88c86403fa06938c427658b52413c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"604a165e1c9f2a9ad9d2459ac361311c73842cfd2b90b7f2f5666c691249ebb4","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"f5dc6095e432aa29c47e82603cb9564458465c53d674aaccd22ccaec43a07e71","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"9e927bdcb6eb5957fcbaccdb025e77f32fbe0b9aa69f0a67324e1cbb752ddc47","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"14c11ba2fda8486715b7f6344682414736242bb2e8274da469faa020bae2836d","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"19e59a263a3cf218909883df36532d62341c3c0c35d73514b93c834483a52898","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"16144d3a7f152c2daaa9b461d29985e1805916d5b3fb1ac8dfcf0c5605c6c439","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2482611","recipe_identity":"8a260e64d501acfe57c71485ae753a6489e8d722eaa5eb9a519451987bb28767","loss":3.4723190263726935,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"6a935bf7077acad6ffed6bffc0df813679c6843ea433efc146448d562a331cef","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"3003123e171e9c5531c1ecf603f09fa0557e630c434d832c4417653e3d1a92e2","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"173b8d55241e40d5c80292d00092cdcdd609cbd5b8dde06d36044cc1976c0add","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"d9b407e5ef15a2eab27f222e5ad40747c3eb565e305d12ec8fca46b4ed8e211f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"cfed53f82c00a709f8d0bc9bbd21e9bb80a8276c7817469c70fa4e0fd588674a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"5187f559a3a9df801c3c7a9a0c3bbac1729aeeb44d505f31790645cfa33ea838","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"31e78f6a711285e1056092eb58ec305fcfb52ce09b4309cf8a2b033eb1736810","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"e0b8478ba8bafff6c8a8548c5c3bf6c1fe03405992f334744613d15c6bac7a57","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"8521f13b41a520dd788666cff07200ca057a43cd02f2de9a572a3d61fa38856e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"1807832c70d5d4dbb9c4cb5687186812f27e862ce80d167614b0a363b9eadb04","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"41c79c67143909a3843240a4b9a7cc523c96dd227a3e0a3318e7d17e291b50cd","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"0a5aa79c736787b79c17a896110908439c0099c4fb29d401aeaf400748b95316","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"9d4d78014bef6ef473a2c541eb48b162365b77c42a43975cf288e6e436fb9472","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"3001eb4761e108d09644d91190f47ef89af33c0adcab9c32adc60a5d4357696b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"e7b9bf7315a6fb264f37a7aa9367b8c3b962ac37c4d01cb93048911a8f7a1cec","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"d25b0ffb0a862c3c2a23636932cb7717868c17979a8cf862ae92c2d4867b1ff8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"fbfd3ae7ede9bc903ac693ee00f35a1fc3dea9a82baf4729709a700f9765e6f0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"95800418772b4f1463cc1f624d86554e395ab0d8369c0f18291b7e9e385f28ed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"1dde115b1706bed9e6ca68c258870e22ce7359346bd33980fa78ad9365968a4b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"f8eeddad9d0165a967f2609f261578ccd5ada964767470890d3bcd5a26c2b666","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"73eea9e225b7b21b2665de8b7b69df84bad666b237757c756f5bee0a8ab15fad","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"e2566e851b7a4c5b17f5d438993befe0b0a52fd285edf2c839108a989e6785ce","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"a2c32d634ae898672a5ef4337bb5545e6bcf2060446c5be72eeaaf8fda78f56b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"51cc55562d3a56790ff8c8bde287c042a9500638003480a40e6343664054faa7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"691be2ec0e45ea7401d2d1fdf1685976a49e1d2d9b03b437a0c4a2a4ab4aa53c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"7c9ad8490a78b2b51aadc9b0b306cd550962084dc8a9ede2f42650a93444fc1d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"d9bc83cdfc00843fc74b975df41e780af87db30d90cbd377aedcd1f25f2ac2e2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"9a88b9cf27fffc25a6decc00addfdf923bd720abc0c3f95d94b5b46f81c86a7c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"0775d7b2e0358aa6e82bbda571d7a45e2d872df20c52f042bfcfd1219449c0c9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"b9343cf26063e857968d0a6245c02cbc665bfed72cb88fe3e7b1bdb77d77294d","bytes":5315}},"requestSha256":"1bc048456d0f64e8542cef1fe1e884fd6446f3f8e341f8ccc3f74cf79fc93430","metricsSha256":"ee0cba6603992e6bada9579925d8c2bd12fd3d3ad920c6482040273cc0704e55","continuationSha256":"ed94541c33b2f04ebcb68acd5d41fd9f6b6f86c0aa8e002c0c9c14ab80f20905","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"73eea9e225b7b21b2665de8b7b69df84bad666b237757c756f5bee0a8ab15fad","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"e2566e851b7a4c5b17f5d438993befe0b0a52fd285edf2c839108a989e6785ce","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"a2c32d634ae898672a5ef4337bb5545e6bcf2060446c5be72eeaaf8fda78f56b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"51cc55562d3a56790ff8c8bde287c042a9500638003480a40e6343664054faa7","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"691be2ec0e45ea7401d2d1fdf1685976a49e1d2d9b03b437a0c4a2a4ab4aa53c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"7c9ad8490a78b2b51aadc9b0b306cd550962084dc8a9ede2f42650a93444fc1d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"d9bc83cdfc00843fc74b975df41e780af87db30d90cbd377aedcd1f25f2ac2e2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"9a88b9cf27fffc25a6decc00addfdf923bd720abc0c3f95d94b5b46f81c86a7c","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"0775d7b2e0358aa6e82bbda571d7a45e2d872df20c52f042bfcfd1219449c0c9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"a33f23ae99693423a4801c08aa5904ff40d4208d48fa41494f1d431fbf3cdbeb","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"31081c272aee34a76083bf9e7bd699f7687fa51b1bd59f696bf40303962ed24d","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8a260e64d501acfe57c71485ae753a6489e8d722eaa5eb9a519451987bb28767","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2482612","recipe_identity":"f220667c2cebcee4598446f10850c61c6309e2588f4e9976e2d6697446bdc424","loss":3.472421223193334,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"a1adfbc6e8f2a16e91ea4732a510dee38e4bf0a40fda05e1b0cd19a47960c918","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"7c5a8e89820afe66c7fad3faab11df24dd04d35c3eae9b6f4b8f8d12fad97098","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"1eb975e5c4c861d8bab62f821b8c15e3aeca4872418cbd2699028cc44995a8e2","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"b41bdf616798f786cdbe63c359a528416dbfe6cddfb191d281a22825d9972351","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"66c7f940f15941e65a489814222175262f60cb3356349d84f5a18695fb3c0006","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"ea5c63304a818e0ff6d3c02e44bf8a904a20610dc144e499db633b37047a64fe","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"a78e39dbee026adc2dcdedc2f96482a1908bb48996b4948bac0a7cb0554f2de4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"cdcd34d509782ad0b1f1edbed91fe17d06cf70cb4edf79ff4d24b9f8e79d08e0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"c277966ee20023d16664dc6d8d3e0ca852a9c28e87fe0de3eebaad0e5c90d98c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"56c5d46d14466b9ee30020a10768568a3eacdbc097f1e5e56e2db4faa024dfd2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"a0176e9de8dd1a04d7c2c5a995a54f4341dab9a838b6c701f85750bb23bdec06","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"e3f8551385660c8b912d2a68bd28da4f8f658cbcdd25df66ccc29ef0c6ab2dd1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"9a1af2575d8a564e45a6e2f48fe283a4f0dd17d08ecf5043cf168f23198107fc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"d3030a27f0906cb254744bc3216c836c96bd7b8ce4b889777c3a3877d6b6f879","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"aa5ec237284cb298a3c07e4aec098d16f4e71593a1dc0daeecbfbedaaf524c80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"86b5603fe1135c26a21de2e337dff880a4bcdc4b6f6b686e3153aaeae945cb09","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"02e6bf341529294b5dcc3ffbe66f8523b78ae224959dfc7f9886267992be35e5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"867569e69559067eb979105749c1a93d136ff76924ec7ad5a80bcde6aca58b21","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"195affec030adf84bfacd483649c0606c8bb57ca8e157dd1c081a43cacbe0913","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"f9c5d1e9498be0a6903edfd5cdbb2cfaa016ea9eda703243212bb6caa08bcd76","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"57aea5ff7257f308201eef4b066a4fd53b53f8119c79acbccc93517d1538d0fd","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"26be4bdd90bb27001ae955985e3edd8530ff500453a61126562874b7065b7333","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"9adc4048a1c8191770c13ca2d99cb83cda4db3d5e708a873c6668202ba631413","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"7623676d5b3d974e94bffe533b79e9a6e7d6548b40aaace015aa62ebd1a0fa62","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"9e58cfa58fd3e4890937fc325c51081680a9ada6ea5f1618010320084ecffa39","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"eb1bf44216f6243cc97ddea7be28ce5c871b4506daa4ba7632d5b1ec7318885a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"e53d827809494d008bd47370082b34a584d530947623555b42fb851de3902e3a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"c895311708f567e1823f1fab6a4b8b35dc5b4ee5bdad23f49d982b0316d3b657","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"ddd880dd77c6193e6de557c9a510453d30c937129ec15b5be95c39225d97a69c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"0d0a843409fceb2fe808557851c5b514c5496a0e3654c6a16de6414d08cf65e8","bytes":5339}},"requestSha256":"bd9441e61425254ac6a0f4f097d1e9d6194afbf9d55a7e2b80b25ab87b0d648c","metricsSha256":"d1b0db63785d980f07865612cd0dc11daa582fba8e0bed50580751bf87a3033a","continuationSha256":"6fd05c229a28b0d838a4c2fc796adae0b335917dc3b7825fe76a961b6d1a8d8d","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"57aea5ff7257f308201eef4b066a4fd53b53f8119c79acbccc93517d1538d0fd","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"26be4bdd90bb27001ae955985e3edd8530ff500453a61126562874b7065b7333","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"9adc4048a1c8191770c13ca2d99cb83cda4db3d5e708a873c6668202ba631413","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"7623676d5b3d974e94bffe533b79e9a6e7d6548b40aaace015aa62ebd1a0fa62","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"9e58cfa58fd3e4890937fc325c51081680a9ada6ea5f1618010320084ecffa39","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"eb1bf44216f6243cc97ddea7be28ce5c871b4506daa4ba7632d5b1ec7318885a","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"e53d827809494d008bd47370082b34a584d530947623555b42fb851de3902e3a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"c895311708f567e1823f1fab6a4b8b35dc5b4ee5bdad23f49d982b0316d3b657","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"ddd880dd77c6193e6de557c9a510453d30c937129ec15b5be95c39225d97a69c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"d2aa471bc85e8c4e4e3c9144bdf838d284ab4905d9ab26c14f507e6072123abd","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4bbdc4414ee734a474c5d212a885011b38ba26d398b7ecc29b6288a363b62ab8","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f220667c2cebcee4598446f10850c61c6309e2588f4e9976e2d6697446bdc424","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2482613","recipe_identity":"c28fc938b8e56f9bd5d55fd60708f0da25825703e15fc4e8e7b6fe34c75406bf","loss":3.38850917424433,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"55e9f9daa8080cc453a9270adc097e194ab8a102b78fca24e089c23cc090749e","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"7e4115bc064c69becbf35f0a9d031924d838b7d0256a4301a76d8977cdf27473","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"6e37983bfb1301f81486ff14c35ec10f12b47c04d321196ad735fb93b08da7e0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"5f3b23467d01b3c99f6c29c62800033f317f2b9c6374e9c26b11a5158d80ad4d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"894dc42ff39461eb8c5abc036dbc1744515b77a32213841e7d4c94db0722f39f","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"fc3528aad1d77b0fc0e11f2d172ec55b6a901855c461351a44f6482c7df8915c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"99101b2a64808ef34a751625356cb1515d2bb337cd3cd45185b2d72a48a8ba81","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"7468aa9c6c1a2c0eada920040b48881622f7b1d03c51f985c75421e347b31aee","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"e119dd04906153900a211e914f90ebd83608fa4f56063be39b0ac24278eaaf0e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"cdc630799e0c0106f8545770d0e15d7d30d79da853c16a4f5d71ad67be9ab933","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"79a7c19e0d00a30f4fecc8e49e7fa00393e245617a97a3a0690daff9b2070718","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"ada3716eb7ac32ce14ca6411cda13329760503071ebabc8c6f49a7112fe171f1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"26ba7c0430ac496ceb043c286339be272cd1861771b611203fceee2df1dfbe4f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"44719a99f0aab38868cb77b75ed8f219a76cd8d953aa100eda736efb3a1f24f6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"47d4c734fca4ad88dfe44cdd1d278e2dd583572e15a4747ff54a4d2e6d894509","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"6a9ac3eda0319e8abd486bc1d5e8e22aeeb2dc19298be024998592a4300ef418","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"54e8d0f45c6fe009eb026ec20570c1422216f907c075c65e6301e30d8fb96c1e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"2daaa7a3095396f151125951266b094599c50a36f525da908bbadeab396cce85","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"5ab449c326196094328f2e7ea51048573b4f1fa583320619e23364133a7258c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"86d23a8c3f3af5af0d0dcd0a2bc4b3aac744e6e29e205bdddaea9620b6f26e4d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"1d23e2c2b60c8431f004519238405b7da776ab0c4db46739c7b9e0184b537399","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"23c0372b995794b48368e4325cdc5fa20fa92f6b02d54bda0533b4e3ed03cb8b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"c0f0082f924a4d761e4c727d930bac2560eced4a4a42d8255925e558d7145161","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"45b2d8c1e2e08c28b23dfc5113ebdfcb20238aee76f9ac58dd1f8f63015795ee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"f414c7f719e087fd20274a47e5d84ce7c63d8ae343ae142f5a10a56d7de38e5b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"3bcabf3c6163d0bb08b3d7434449b2535440735d7c4a8f864d258cc643a4b930","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"e11167b0c041026eb1f2fb8f0e6d6c6f9586258543b46c9276c186d1feac1736","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"c27a5f3b50e023154033295cf4c3e84d9bbe4bd1d833e95fa5828728093bc0ad","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"06bd206e4ae2c63f319b224b1193b5b82aaf73c2420fa017dafcbcde33bc7894","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"19b6d13257550b36162342680aab82b736caaa97c971817b69b14aca52108a40","bytes":5252}},"requestSha256":"3a9e15ecebbc5f00e72952c5e0454bc55901c76cedc2373c43bd02cca6f00c5f","metricsSha256":"6869bfdf2c4f68578a7837ef837fb4b7f41861a1f526eb26a58b8d38dd7bb253","continuationSha256":"cd9f6b473a28ab5699ff39f044de507b5b699a78a5aa0e92dc52e98198352eaa","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"1d23e2c2b60c8431f004519238405b7da776ab0c4db46739c7b9e0184b537399","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"23c0372b995794b48368e4325cdc5fa20fa92f6b02d54bda0533b4e3ed03cb8b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"c0f0082f924a4d761e4c727d930bac2560eced4a4a42d8255925e558d7145161","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"45b2d8c1e2e08c28b23dfc5113ebdfcb20238aee76f9ac58dd1f8f63015795ee","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"f414c7f719e087fd20274a47e5d84ce7c63d8ae343ae142f5a10a56d7de38e5b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"3bcabf3c6163d0bb08b3d7434449b2535440735d7c4a8f864d258cc643a4b930","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"e11167b0c041026eb1f2fb8f0e6d6c6f9586258543b46c9276c186d1feac1736","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"c27a5f3b50e023154033295cf4c3e84d9bbe4bd1d833e95fa5828728093bc0ad","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"06bd206e4ae2c63f319b224b1193b5b82aaf73c2420fa017dafcbcde33bc7894","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"0258382c479c0afa529ee6acadefddb050114b490a72cdee68f4f5d502fd4313","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9685eb3d5e8d5c50551f5eeee2731bd7263f0af96e8469299c84c49ac7a7f32a","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c28fc938b8e56f9bd5d55fd60708f0da25825703e15fc4e8e7b6fe34c75406bf","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2482614","recipe_identity":"bf5c6c37216ee4f346976b7c06ac2383f6759a52513e8d5736f38ac428e5f3da","loss":3.396818354457583,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"9615eb4373d1648adce950c692f6abf341d26a5d8c027f51e48360675064ac5f","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"d7bf6fd54ddd4ed592c9f441ce3c1f8392c28379b4c6b3df2b7eb00a7ec15e88","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"541a0f420e03ac515538a4c5c14a2da9831b7cf3a187bc8a60eb051b6790c79d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"7ecc6e50914c1abc0f8636115b065511eeecb622a6223d2500a9c366ed355286","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"029fcdb2d5043cc670d15812abc98c36f1d52f874788a429c9a1c8b9c9ec10d2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"5816441ca8cde888a606a7aa539d2f7ce9fd0871949b5bb908f650fa134e0910","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"38da11907b467ef9c46f8aa9d6d1dc5145a8b1ec0f0eec7762fea6b99dd1f9ad","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"d53da494060064571b0f4905145819324e16c353f9563db1f03a1179ea4ab024","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"e7bbcd5e7b648e8c35a6657ff99c6f79550cfb23986929de8ca7d9d54bf4d203","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"0737ee88ca7d7e8a2d1a17a158da3bc52bb353096f98f0732ade68cc843361cc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"ae5be71e8f233ff6109393bf3e212b9371b02412a6ecca07daf9bd7aede6c117","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"7e83ae2e8011c08296eb234fcbd11c2a428ef74d28e5db9f356d16baa40d7e31","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"2d17d059ef49cf982389bebec7e59c0f8def0c9ec421d98086e6e110ee2c79b9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"5cf984f53a27247bf902a1c389ae298a1f310e15f05061826c0657ec7b8e40e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"b3e28d8dfebc42d59f4f780a0f199c9368d91e0b9df6d65c31c1c07e56e8eb92","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"514d8aee6f1aca58df22e14a1bd4f823ddbaa10086472efdc8e0ee333a429e01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"aa0f4480f46678150945baf18c499501c8ce5aadff55ea3ff6bec5a2143c7b3b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"2100db948ccb37c1e396d1d08a60ed82fd9caa026fdc921b2c138f4c17217e25","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"d0c940328dfafaad39c00943749ef5f650cffeb91ebfb2f4057f71cfd5db66e6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"7452b1c8355a38b12107ec1c674fb7ff3d160a948b29ef3c04bdde9209b016d6","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1eaf51e19ba7fc6b4348c45c9910e18dc04e22ca2595e1de4cbb9335c234cbe0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"57231d2193a67fad7d2b17112636994698a88b931b6a9d06faee930e4fe9b399","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"1ad2c95b9fa684d620f237840051d9552cf0b3a99f82eb7e470045c155220287","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"bcbaa9c12706e92f0940920fde03b42ff3b3cbccacbc1a2618420a8883cf8edf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"3f765e991030d0b739a411566006663ae5ab7742ef28310eb51aee2fffbf6f07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"e56174fd8693ed7744159bc410f37841a66424acfd3de09aadec256cadaa905b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"3a7dd06abf27a6ae32a5938c99865d79f8553637cc17a88fb72109c5310589c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"798f5febbb37fc11ac967c6766d30da545075b55dce3341cfcdf62c21d4cd310","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"419af95fee81e750144f357e37a962f375648d382b565dafeccae3b41ee71fd1","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"ce959c8a460babdbcccfe7c980cb64507b30e76d60f637f34964dd48d2e80a05","bytes":5258}},"requestSha256":"46268d77d20bdd3f144bd910778e2af6b4d377509b922a22f299e7059a914c96","metricsSha256":"49c1f9401e9449127b00cc57be1d5b65ea3778bf877996abcc1903d57cd39e73","continuationSha256":"3ba22889ac9ad044c3d40856a09c9e5c0468abb99b0d45f3b164d53f57baddea","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1eaf51e19ba7fc6b4348c45c9910e18dc04e22ca2595e1de4cbb9335c234cbe0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"57231d2193a67fad7d2b17112636994698a88b931b6a9d06faee930e4fe9b399","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"1ad2c95b9fa684d620f237840051d9552cf0b3a99f82eb7e470045c155220287","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"bcbaa9c12706e92f0940920fde03b42ff3b3cbccacbc1a2618420a8883cf8edf","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"3f765e991030d0b739a411566006663ae5ab7742ef28310eb51aee2fffbf6f07","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"e56174fd8693ed7744159bc410f37841a66424acfd3de09aadec256cadaa905b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"3a7dd06abf27a6ae32a5938c99865d79f8553637cc17a88fb72109c5310589c9","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"798f5febbb37fc11ac967c6766d30da545075b55dce3341cfcdf62c21d4cd310","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"419af95fee81e750144f357e37a962f375648d382b565dafeccae3b41ee71fd1","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"9ff93e34b518d427d1e3cbedf6b8857eb954f0e1dfc8f403fb95a3bc03830a58","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"208df82acf96e75064e44629ea708657712f17990880b3896646b07865841da5","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bf5c6c37216ee4f346976b7c06ac2383f6759a52513e8d5736f38ac428e5f3da","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2482615","recipe_identity":"55cae209132436c2ebf0712742eff9cf12cc8f43b16de75c683c6419d49c6661","loss":3.3946147154764614,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"87708825f6707e317ebd100b4484dc501339d65a6c214523f6d9c192c625e7de","bytes":881},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"2b70ca89bcb4bb5a2ac5f8f5e99bb4d882aa7b28b9ba9965ab82f0daeca5b04a","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"0c9b2be66dc2a5b0387c7687323dce2c19f15effc4367b30c6187e12e5f46b56","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"f6b346fffec8decb0cc56571fb15d6d4c52370c726b6929920eeeb5d0b1dafce","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"714eafb6a199550973c3b2643a3eea845fb7ce67b71fd6325f5480d4b29f9d41","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"6d2e53d9f6b22d6854beb97f3ce4225d2b0b7b780249e9584d3438dcbc1032ee","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"f24e0a08bf2a30b232c157a322fdbb0771939436a92b1fc71b37775c89ab6ca9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"82360544d54663a87765567f84fdf9393f9487cfb849ee5dd59c4d9e3d28c590","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"966fd04272dfb659df7dbf90f8f689ef26b9064e7b3a17ae97a5c253de2f543d","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"760c6e066715cd7c204ef6cb42578aad0a25d69053301a3ca0e94e89e7f06640","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"dfaf2efb8f926297be139515dbc497132f534d5275d6f786f1213abe0a604dd5","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"73bfbc44fad108143e171bbf97a95b5f91eafc179c7d29937a65cfbe9f92f8a1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"c90076d878f9e553a3b8fcfda96a825d28ae20d827b83cc6babd2b231aad0e9a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"26b99350b83fd3730a7422590c59e716e3423a9beeade70178962a3df20e9d2d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"ae9082338f5b65d0da1e517e9e30ab74a35780a2932ee7403687935c78b220fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"cd3152d6475f26ea3857e543253f0170ffe150778a34c8e6ecbc4df584e83bf5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"59d8d27a13fff700aafda8a333016e8f32778752d201dc9609ac6221cf691219","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"a97a8e194bf3a3449ff98f0d29e284fd707c4965c8a0670ff4663ddb32f4122c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"187c4ee253da9cb28c9bbd4801363792e72f686e601a1e47ced24b91614a9e1d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"2afaf285e788c234e54e18f35763b2338b6a5303a0589fbff2a3e3d1f7ce3b77","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"de2d73618f9eea0e057e3944079e217f4a12642dc0b66853749efbd57cb6173c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"651f2b711b5115bba3cede325fa34d1b4d3086e6a5132f645c9fa90385a206e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"22710dcc178a9d673fc0b769e04e191b290189a80532943daecb792a3e3f059e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"043b0c2175117b84db68b0ccf1a625eed34ce4b769a314713cc11355ddb7f891","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"a09b8d984213ab4209c8ce650ee1ed37916cc584c50c62e9ed88bd817510031e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"b34c5f172b4f093499c58ee6030b690def2fd9a023cc04bbd51cc4aa4357c238","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"a56b4b74752340586b385b92d187dc7f91ffabec9e818120562130dbb9316c0d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"e4f685f754ce7c792af53670c127b3348a85ae89d65d1e2a585ed17cb9d57f19","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"4437c1999aff1f0f0cca4047b249c217cc5793609bbdd8faaf1c4b56e96fc139","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"12a16688aa2d289f4e3c7cbed2bc45c05a59af14223a358496a1b8d22776d7b3","bytes":5343}},"requestSha256":"e55ae0a6daf77730ff565fd03390152e39ab6447fec09932494d4d7ce795ebe3","metricsSha256":"e86103e0f623fca22ef7405e338e7a65186f399304ac940262ba03264568df11","continuationSha256":"51953b356a0047db11b44bb0e0e4c211accc74f151efec1b37d8e8fe1c606caf","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"de2d73618f9eea0e057e3944079e217f4a12642dc0b66853749efbd57cb6173c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"651f2b711b5115bba3cede325fa34d1b4d3086e6a5132f645c9fa90385a206e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"22710dcc178a9d673fc0b769e04e191b290189a80532943daecb792a3e3f059e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"043b0c2175117b84db68b0ccf1a625eed34ce4b769a314713cc11355ddb7f891","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"a09b8d984213ab4209c8ce650ee1ed37916cc584c50c62e9ed88bd817510031e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"b34c5f172b4f093499c58ee6030b690def2fd9a023cc04bbd51cc4aa4357c238","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"a56b4b74752340586b385b92d187dc7f91ffabec9e818120562130dbb9316c0d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"e4f685f754ce7c792af53670c127b3348a85ae89d65d1e2a585ed17cb9d57f19","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"4437c1999aff1f0f0cca4047b249c217cc5793609bbdd8faaf1c4b56e96fc139","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"d1ddec220cb48686640d447640a108614bb84d7a3c660a6bebe1825257ddd348","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"66124a8942fa1e4b467eb3132973809dbefde0a4f93e6399462369a9a8c3dcc8","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"55cae209132436c2ebf0712742eff9cf12cc8f43b16de75c683c6419d49c6661","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2482617","recipe_identity":"264866f8998f2397795ea697994cf8fbd621da67605308f631e1ddd2fda56873","loss":3.408977548963366,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"5783681f58b8e516e863a61bd7149b13dad6b855916ba128d9ada3da4651d5e9","bytes":880},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"d74c46dc663df9e6f7fe5f7622636ce98677beefdf6bc4d515fe138ccfc437ac","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"7ce75a49a91ebc0a173aca38d82fff21847ae450c3a260e6273d31b71d7e66e0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"33e8674b1f9ada9722d3f2e7f6d08bea738ffaeddfc752ccfba73fec2e74e34e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"939350c0c4eefe069cc2aab0c69f421ec48a142a887e51a5a4f5373deae387eb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"afd7a9da6f0627433fb49e95a4b835d3137d074eefab1dd97e9ac4eea7bfcf3b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"ffd6f3ff7b133d0e8198e104dbb30ab7da4e0306f7f9ceb87a86d2dfd8130ca9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"fd1f3f14cd23043c5368377c1692750e732325b0f803042f9a96e9808a28a404","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"ec4d14522abda302422d065854ebefa69f0e06d92025e1cb2bd379e58112c8a5","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"6097ae1f6cdf52163b84a496553dce3a1f92d20cd583d92e2fa3dee62003fe17","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"ba14813fcf59e40e9c150826ab3c75a3839852a3004b8394c999b2fdb65872fc","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"67958cf0d829bf44deff8aa036ecae19ae35aaa2ac8e2b5cc7c31682f2e9bf34","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"5686ae856dd79649d9f6021d1726c814e2eda1b2c2274b8c84cbaadb860521a6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"c6b844657d406daf227735e086d49a2f7b7c141fb9d07c54a16b388c5bae8496","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"5fa76c530c03c4d979a22d7ffd23ff1d870dfa1d79a1f50d25a7319e81a97d04","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"c974f5ad295ec0e0b9202cbadde0875220a1282cdc0d5072003a883558d220a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"15d85d8130eb0bea28bbeada94560f2ddc8d7a758c172b61b6ce9ecf13c32ab4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"89cfd69d568268b23e1b6fa4afe43a46f938f5d767def88a3edd5ce4a5609f15","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"4fe709115225b8ece1af20d4e8ac71ef8ad9192a692f6a8a517268f3e1d9f399","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"14ca2ad52cd5a959352fc742e6455b690c4e42324053056426ea43e2274db096","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"e3377a7b418694e39234eef6e9050c72fbef6a4be2b9ed3d5bf4914edfb6b3e0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"c578949c2c7bbd493686b33e21bbe0177c6483d265be12d3bdffc6f30e06c9f5","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"0a64d40c3094b869eba9ce9adc406ecf056a5a29585e00539dcc791ec106ee87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"fd18bf50d0e0a1c4e8763aa5c391a78e8b8310287580da6359f6c948224e3f0c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"cb78b791c6e3890d5a2ed1ebe9b14235de63b4c0889a0cb2e2d0ab9718928e47","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"395ec227b5aedfc59be1b91e62a92715d6cea94f5e446a94e8a7c4b12b3482e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"917fa671b6b9ea348919bb89de18d190b2a209ad0ed2c56e31dac793d3876405","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"75ded4be692f859fe3b6c96c3ef461edb4bc89f3a3fea60393756c3481164c2b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"300552b8a5f8e87533ce73dd831979931e59bc0a739543335bd4002c0030ac4e","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"b084e354a2d6e4b895ef4841d9a5b71d10966f1de90dcaf43263428fbca6f871","bytes":5343}},"requestSha256":"6a3d92f090633c6eb751ac24cbb815f10352d28233ba57073697fd91b343a2ba","metricsSha256":"c94fd864d6d2eb1c724a410da0e90995dcd3dab56ab7d60a8c2cf39819f708b0","continuationSha256":"9c0de28a16608458206a06432e25c2d328cbc9754ee8d40a37cd722295449f47","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"e3377a7b418694e39234eef6e9050c72fbef6a4be2b9ed3d5bf4914edfb6b3e0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"c578949c2c7bbd493686b33e21bbe0177c6483d265be12d3bdffc6f30e06c9f5","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"0a64d40c3094b869eba9ce9adc406ecf056a5a29585e00539dcc791ec106ee87","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"fd18bf50d0e0a1c4e8763aa5c391a78e8b8310287580da6359f6c948224e3f0c","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"cb78b791c6e3890d5a2ed1ebe9b14235de63b4c0889a0cb2e2d0ab9718928e47","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"395ec227b5aedfc59be1b91e62a92715d6cea94f5e446a94e8a7c4b12b3482e0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"917fa671b6b9ea348919bb89de18d190b2a209ad0ed2c56e31dac793d3876405","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"75ded4be692f859fe3b6c96c3ef461edb4bc89f3a3fea60393756c3481164c2b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"300552b8a5f8e87533ce73dd831979931e59bc0a739543335bd4002c0030ac4e","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"aeda91fed7f8f7c30e0873269ff1c8d54911605684402460e4141fb5105b0912","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e92e02e398e7ca62a11f30e1d904a947e893219e583a2316e7da4f1863bf7774","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"264866f8998f2397795ea697994cf8fbd621da67605308f631e1ddd2fda56873","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2482618","recipe_identity":"05376aed83bf802b53f7cd41f3660d9a87be4c3e2c01855c72bc02933444dc73","loss":3.413505536810266,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"efbfb1ff303896a0d36ce5a293dac2d80e0921b107169ce460e63ef11b5be1ff","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"12f77555fb7f7fd4632d9440fddba76f987137d582c9f8ea6d02f3cbeb029929","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"5d33b019c981fc7fadccacf06c4ba079cc011c1fd06a472bb9c75756f7438e63","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"1660434ef4f993f6fe96feb3d698092687f78df649643e06c7587ff13d4edb4b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"75754a066807a34b90c1f14e40a5fe2de13e692f39469cb7cacfc7dc21fd5387","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"92637f9c62d44e176ca8bb8e6da33deeb608ec2e298fb57e704c420e34f0d4c0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"9e650d74a77fe51a109ad4f1016d7e80f6e7a7c3f98e6cd14ddafff07e16ff9b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"c96bf8d4ea3d53699f79dea60672f1487a8ea1f5bb91b65fab424e5172c247b9","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"876e35468d5e1eea3a81f4fde854feb6c0584f110c9b002c313c9a146a98ad86","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"5a4ac4ed21b2ce33a037d676d547824376ea5c439d544416b70a6262b457d07f","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"55a892035dd349ad21340e9651bf21b46b45e23cf4e97365c2e5c9dc944faac9","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"b23d82bfd54b140d644d27dec68a7b7f1d40a6f5f2e04f8b2da6afb204bba118","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"3979ffa739a7b54d1f61ab66fa5ad7efaa659cef9cc900ab307db1e01689e582","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"cf55855b3416d24af7f5baf6a99ae750e53f0c3b9dd9e51f4ecc767d18c0cdf2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"9c8f8d7d222c0b04681701e5af20bab64803c65cd78c8acc6331f535f374efd4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"a792ce4712f476fe07494fbf4139e07dcb536e3e55ef7c2101a32f95af55bc04","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"0df88799f7e4d7c65a0d563f8554630956964e721e2ebb53970501ca7111a184","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"2199ea39db8ae978701175638534f1289dabc850aafa7712e066b187893cdfc0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"b405f7a07054132f38f37ae2172000b2d953f30f64193bbcbae05de0f1b69c4d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"8a059aa7c1a667c8c62735544b92e04b102d495d1cbc4612b3ca028270e55d03","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"b2765b30e10868d3c7596a6801fe7a00b4f511b7709f8c2affc68f22daf97b1a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"fc1291c5bf6b1e37571293edb721d396bca71eaa9eda61962482885d253b3817","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"e123f49d1cbcfae8228c2b0aa143c9bf92b2c4d6aef650eb41ff5da09620b983","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"21cfb260bddb6a46e9db464f50210ee1ad930ab944468e73ebaf572f8c1947ed","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"aac3d1db910179fd173b467169954912bbf9eba5e07324248c547f8c6cc16e81","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"a9966108ab87b16052b9ebba1685ed3800dd5495235753493782426f4bdc9f8d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"a1febc4ed22864fa36bd914bb9ab14060889f128c9a0da1f6bdc49443ec2fc68","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"9e27b110a373269808ef699246341d13c413baa13cf8abe9be4f8465f1905e64","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"0068e4670879500f96002e7d8d50aa5a6625b718139b02a0af18b92aa1203390","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"e2d56616c451d04b6f8c0374d3e16ad9f85750c5e521c6f9f731234d72a955d9","bytes":5338}},"requestSha256":"e6aff493b508ac9927b2ddd44811cee34fab01b2aea66b0f6b33aba556227c3d","metricsSha256":"8ac2013a929f0a1d40106935cf600a16b87b2691c4208450f69c40f75a3eb872","continuationSha256":"3585d9897c7404748d80ac5ce471566dd6a6a86812469ad170629a725e63fddd","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"b2765b30e10868d3c7596a6801fe7a00b4f511b7709f8c2affc68f22daf97b1a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"fc1291c5bf6b1e37571293edb721d396bca71eaa9eda61962482885d253b3817","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"e123f49d1cbcfae8228c2b0aa143c9bf92b2c4d6aef650eb41ff5da09620b983","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"21cfb260bddb6a46e9db464f50210ee1ad930ab944468e73ebaf572f8c1947ed","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"aac3d1db910179fd173b467169954912bbf9eba5e07324248c547f8c6cc16e81","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"a9966108ab87b16052b9ebba1685ed3800dd5495235753493782426f4bdc9f8d","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"a1febc4ed22864fa36bd914bb9ab14060889f128c9a0da1f6bdc49443ec2fc68","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"9e27b110a373269808ef699246341d13c413baa13cf8abe9be4f8465f1905e64","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"0068e4670879500f96002e7d8d50aa5a6625b718139b02a0af18b92aa1203390","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"5149fc493417d96429d09c714ccfa13270f29498dcfdd3df7964263091710182","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3e3b7bef8628c4bb881c7c66752c238287669d6a9ed63e9f2c20c5cca951b854","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"05376aed83bf802b53f7cd41f3660d9a87be4c3e2c01855c72bc02933444dc73","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2482619","recipe_identity":"6a48c879e3c73bd0914fc1aacc6b292a16299252e07e96cbf0019d222e7ac1df","loss":3.4067449574333812,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"75b5fabdfe554c1f85ed68d21a15c40883b002709e0177012ffe086edaa205a1","bytes":886},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"e1da7f19b2b43c4132644acf4bd92f48b3723e063299c231030e22126a60fdc7","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"e2094bdcfe931b070b3d45d349abb91497b78715f9d580290522151e7ae0d45f","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"007fd4f9bfaac7aee6dc9d21153f682aa2a1c4d6871d6f79d529676223f4d261","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"a2f9eb1c9ddb41e52cdcaa2f91686f193179dc4b7013feef5aeb20ec1d6ed469","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"13805052fb3f37e0f829ff9e2244df56df617d7ab807a577c9e4065cd29d1973","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"aefca8afacce3f11029ace2721b39834534c6b3aedfd965919babf6c72e6565d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"4c6822711efc60c08b1c7d90ad85c9af23190311d8bad26afa62912268218fbe","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"c265e5622dfc4be03a8bac70f6767b5cc03ca2022575cca7ba0524b2d26875c8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"f0ae9d67d197fd76883836a3f0e0f39bd6cb243ba61e3c327b3bff34ce2e0e13","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"d7f133187371e2891165f9e60cb59d668c6eb05b5810b1a063b1c4638874ef49","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"65b47a0f8d24f5218b91cf8e28bf7548487253d4fcaf04d59bfe3f39ee370c63","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"9a447f15a7d7e8ddd63a560d06537cc10c86847db5c6c7c277725b7f70746a3e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"19f8fb95a08616eb61f3941c79ba1ea99be9ba8d3bba2325cd14cab64233a2df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"872213dc6c6fbff5ea21293e6fe22b2a03ed45b01d9188ca8b120f54cfd4da8b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"7deca9dfd0c6c434be42e75c30375c56cd023cdf2f6ccd4a6773b77c190db84e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"55c87869edbbfd68c777979793ccad6717a7819e7e688f7591b6a4f4d9d3b382","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"e40d0ea4fc93641c915a0205ae5adeb9a5b998cd951d37551cf34ee187579571","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"5f00173e341baeba254d568ed5dab3e16771cf1ba817e3469bfa99eb7d2e233d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"9cd7083d0ccca26fc5408792f6c1c546f0f470ef23157c2454c43daa5f7ae08d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"5409eeb6cbadc55c7df8e153ac76f4037f99b74756949003b6793520a5478f51","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"ac8ef820500e8999cddad8c7e5e4d69cabae0951f29ebeaf868f03fc93e6558f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"0fe2003a43d7b6e57260231ea3351120e8a15a4f77e50250f289269535b657a9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"deb75348ec677ee3f7397474ceea22c9c6735c1f040ac2e6ab57f32d6fb72ddc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"bb384dcb3f3448a798d543e81910053f6cb33a01273f4db5450abf9d8b7f9d6d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"d8a1bac51772fb14d968fe412b90750a59f46d689a03d3f32085ab6cc37b4310","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"3cf480cc859dd3534cc2000ceb5fcafd837e782094e72324e42366ab54d907da","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"e1ba070519d4c22aa6e945dbec015503f22decd7c883f98745cb8b5c14672fae","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"a1037af21826a2823eb00dc5c435bd05f2a32f6ce07e5045bab4f649e9019ca9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"9d33b70f070a322e31e4acb987727b3db0beead5c1693c616f3049494cdf80c1","bytes":5314}},"requestSha256":"7cbda5ceba7fe9829da1abfe342bbec44a1a9729b9c7b07f42e3e2047fa74744","metricsSha256":"8f24d0ba21b9a0a711a57d526f80728b5584cf43fc9652aff37712125a2a5cd2","continuationSha256":"c6bfef20a7880c0cd367ddd5a9e61ec4da0c8ac8c9ddffae545a289fd1618503","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"5409eeb6cbadc55c7df8e153ac76f4037f99b74756949003b6793520a5478f51","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"ac8ef820500e8999cddad8c7e5e4d69cabae0951f29ebeaf868f03fc93e6558f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"0fe2003a43d7b6e57260231ea3351120e8a15a4f77e50250f289269535b657a9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"deb75348ec677ee3f7397474ceea22c9c6735c1f040ac2e6ab57f32d6fb72ddc","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"bb384dcb3f3448a798d543e81910053f6cb33a01273f4db5450abf9d8b7f9d6d","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"d8a1bac51772fb14d968fe412b90750a59f46d689a03d3f32085ab6cc37b4310","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"3cf480cc859dd3534cc2000ceb5fcafd837e782094e72324e42366ab54d907da","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"e1ba070519d4c22aa6e945dbec015503f22decd7c883f98745cb8b5c14672fae","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"a1037af21826a2823eb00dc5c435bd05f2a32f6ce07e5045bab4f649e9019ca9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"a7f056f13874b5c31fd674701072807436714207c2df5a4b4a887d20d871f0ee","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6028f48984e92aa1337ed658409ab8fd39c48128b96aaf69adc8615b99a9d14d","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6a48c879e3c73bd0914fc1aacc6b292a16299252e07e96cbf0019d222e7ac1df","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":120,"attempt":1,"job_id":"2482620","recipe_identity":"8d056b802156d1eef5ef9ee9728ac889d0e2f2f724fe77f7990084c25263c4f7","loss":3.351924329591714,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"22ae14e2a109efea50cf05ef6a2597916dba9594632dd7fbe11af203055b5098","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"efea8fc7f371c886009ea833be1db575d8411a14aed238b4f3a5ae3b1c6f9afb","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"ca96dff5bfc7cf448d7f7d9e21339ddf9fb261398b748f0e28c6db801c5ccce4","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"32729a023b66e03a34136d9c4827ac95c0d0d9dd07e442ae4732a545debe4941","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"3f9a6bc6802b655845da9b8a78c1d760d073cb1a80fae1b6fb4dcd9275ed60d3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"709a59aa53b8bc46cb01184af5cb991424de3438df739e8ae1a287db882c164c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"eb36389aaa1b59a5a35fe9d068e46fdc8e586bde4f123bc425df8d12865d4abd","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"9ca20bf6ac578056125c8a9dc215a899bdeed186382f396eaeac5735ec1c383d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"63f5af75574765a06fb030e32cff04a3c48afe8d78d06aeee3c30e54c3294f95","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"992ec79b0e460f81c8332ad583765bd6f7874953c26ac7998e14177f6240c156","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"dd482c29f5e2dc58eabce50e61315a4c75e41f1eea160745b68a14210893acc1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"05a19adeb45a616985a272a6519861fd8093f33d2b407803241f4d3fafb3ab59","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"384cf9c17288168a858ed8523df55aa6ca2871e2d978a38fb2910be30c19c235","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"eea86e38756b4a88e29a2f92369e40ae7b4975377fc1efb2eb08d1ceb50450b3","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"2645edff0bfaf454271b9a0e393f3df1758fbd12e06e56c2ba904166ad0d93d4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"a8aefe522d89d1db90ba0acf641199b641f616a8901e2e2a9e12c1b0a0b8481e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"36dc67ac3f50de46550b343bbdd3b0bcc43cd81badde0b2625f8bcc7ebed76bb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"25fbe1855dbbc4f4cab79c9940c8994b4d17af907401fe1a21299a4ec8dcc3be","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"c09575bd7dbce56bbd2b6516c83684917ea8774dcbdf622800c2f9fba4933e44","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"e4f85bf9b9e1dc56bc78a18620befd72725bb5897b4c953ce17649ee10b3ab90","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"2a9c90ef3e897ddfbf36c2b62efb7292dacc278275f37d790fb756512be08c9e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"2532825d2401213902d6d017a40ed86eca84feec4bb7ffafabffb4a0c58e9727","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"755125eaa93f1305398ec4389ee158a6d7ae4acde2c3438869e8e4a057c6976c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"8cb88ca384c8a14c13a3945178c933be767b1940e5d3b61ac40daa93c9818cfa","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"3374e3c876672f7a542c7606301142aba79ebabaeec9b0e1d4448b224b673dcb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"8556484f6a3df53aa0b82cd812897a2ef90e4f2128f82f86f552ccbd5df9e9c2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"90d24a0085da404b3fa193f64402c16931401a5050bfe9bde54fd69210bb432a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"8aa9eea638ab1eed60e565c9b8305d4917e9f74e3accb9f134639a3ea6a19776","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"490c3543eaf1c30944b8838f5efcd242b261d52655c8ce79b301364144b0d3ca","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"d7b6c025c7fae2e4cc3453fd7a778ddea8c0a24a8f2b7e68b9a98e86c7f5c5df","bytes":5314}},"requestSha256":"9ada11cd1269e7207a0647478afb503c033c8cdb655242813acf4b492195267f","metricsSha256":"26578a94b401e72bf753aa65455cad381c1b53e4eefb7840b8e665ebc0c95183","continuationSha256":"7b623da879a8e97f4a7ec411bc77262978dbdf2ef2613b5e0dfea6a59db2be5f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-143.pt","sha256":"2a9c90ef3e897ddfbf36c2b62efb7292dacc278275f37d790fb756512be08c9e","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-143.pt","sha256":"2532825d2401213902d6d017a40ed86eca84feec4bb7ffafabffb4a0c58e9727","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-143.pt","sha256":"755125eaa93f1305398ec4389ee158a6d7ae4acde2c3438869e8e4a057c6976c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-143.pt","sha256":"8cb88ca384c8a14c13a3945178c933be767b1940e5d3b61ac40daa93c9818cfa","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-143.pt","sha256":"3374e3c876672f7a542c7606301142aba79ebabaeec9b0e1d4448b224b673dcb","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-143.pt","sha256":"8556484f6a3df53aa0b82cd812897a2ef90e4f2128f82f86f552ccbd5df9e9c2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-143.pt","sha256":"90d24a0085da404b3fa193f64402c16931401a5050bfe9bde54fd69210bb432a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-143.pt","sha256":"8aa9eea638ab1eed60e565c9b8305d4917e9f74e3accb9f134639a3ea6a19776","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-143.pt","sha256":"490c3543eaf1c30944b8838f5efcd242b261d52655c8ce79b301364144b0d3ca","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"a80c85b12604b16f7116227ed1b50c27e28d030a35c20f2702860ad75af32285","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"76f519163ac233eb6964fbf33c7e709daea62607161e95d3ee4d204b68e661c0","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8d056b802156d1eef5ef9ee9728ac889d0e2f2f724fe77f7990084c25263c4f7","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":120,"attempt":1,"job_id":"2482621","recipe_identity":"0b86a5f6d8887085fc19a165db95b23f154d99f32e80d2f44a2ba97a063f037b","loss":3.359696544611667,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"57682b90a1789b34dc4c59966e60b2af470fa80c83f00eaf7c61f1b45ce8cce2","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"cb6688543831ae625cdad21e614ecc08dbe4de10cb80a116d743decc30e07569","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"2f36bb80e1c68ad0c8a15657c459f455ca956dafd24863bcb3fdb6083c1fcdab","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"f7c1ae1019c1f530743e67da33ebf6e2b3409d22322c72a185981a8ca3e4105f","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"9d1b34fefd43d991ecc6a84c58b024db288bc6b5e710771b40f8fe5af7c906a7","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"168c8e0a3a66b3f0114f65fc2ce231af062f2b262db11acd6337f93b54e87ec8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"396fbf53428f492a6c73aa266caa9369d65b76365475ec6a08b2fba8daca9ab8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"2a61b44dc47876c80ede6fc572ff2e09f34d895597d347de07d82174d0ec2be2","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"4f9f7a2eb777179c64d7d8692b8cbc23d23b24feb7b17f39a321b4afcf906c02","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"ce5e54fbe304f0f99c71ce615f04e915d86102f86f5a8e638ce15eccca605315","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"78d4c740e5ab6cb6e0127bbb41c7f2c101a74966c733e4ecb5d95d6a48d6eb5c","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"7b0eb71271aa852af4024c1fd577588c656d8e0f2eeb76bc2df82c2dea041d5d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"896b5fde1f4cdc96c346a25ccbea39c9de3792e709b615a065a2a5c176505a51","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"dde3ab2f345ef0a5068e543086bd539d0634024273d796d7474196f288813727","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"49b392e12e761984965e274f600f67ec424a83b3888c5e03c7b33f3120ca2102","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"ba7fd3717320feff954f2e24d3a3d2a637ed87d4a63037eed96914746298f414","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"811c162e26500d5c175553842377ef98030430bd0dbcf89fb160654f44841470","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"acc824203032115b487882493da295d749895fb65fcaa84f05e74d75bc6a4564","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"ab6d6d36330e2b388910fb514d15dce87fadb40d99431e6f306f0686393b927e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"f75ba20242c596dc9eb23edd4cbfc42a980d443dec1e90f6091743d597dc1c80","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1492e8591ea2fc100602d98a5637cbfb0fa3d19408ae6e95c58429df3246b11c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e3f77aee6c53fcc38236fed290efb61fe3d0ddbcc1503bef44ee474ab16a6791","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"85e98cce57a4dd8f11562b462498f5a0c6cc0fcf4444b6ae84188483422598a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"f404a5301622ecc696d1452264c5ff6af87265b3fc2ee88f31a1e028d33329e5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"4bc2700bff93acf1b45353d7e77c48bb1148dde20f73db5588afcefeea260c49","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"fb0d7c27ae5119616826a4185de79cff968d787632657b62bdd0a3b89d853587","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"e7601f1ba9fd632900bdb1130274d4b1e06e2fd1e6a936a9bcb156996e17e82b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"ae449f74fd50272b8ad10a2ee60521d06f3abc5c9ac96d82cf58288453ed248a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"f6da5a05b49d5a521873653ad3216029a12d900637a0b345fb08b3fe7636a4f8","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"798117e6a02cfea96805f21a755cf1658ab0a0ac98d7411ed8cb3d29c477f878","bytes":5343}},"requestSha256":"3f89b1fa5f674f9e30cad041b9f6a3008e50f711f48db7791e9a41275b9d84a0","metricsSha256":"db784cd7d68f44196a8c3e69253db5bad6241347e67db3253f45a6c13b6181eb","continuationSha256":"e5e0b0cf66bca42ee323af1dcc4cf40e7e1f44840a9a911fc9b452428d600157","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-143.pt","sha256":"1492e8591ea2fc100602d98a5637cbfb0fa3d19408ae6e95c58429df3246b11c","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-143.pt","sha256":"e3f77aee6c53fcc38236fed290efb61fe3d0ddbcc1503bef44ee474ab16a6791","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-143.pt","sha256":"85e98cce57a4dd8f11562b462498f5a0c6cc0fcf4444b6ae84188483422598a0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-143.pt","sha256":"f404a5301622ecc696d1452264c5ff6af87265b3fc2ee88f31a1e028d33329e5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-143.pt","sha256":"4bc2700bff93acf1b45353d7e77c48bb1148dde20f73db5588afcefeea260c49","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-143.pt","sha256":"fb0d7c27ae5119616826a4185de79cff968d787632657b62bdd0a3b89d853587","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-143.pt","sha256":"e7601f1ba9fd632900bdb1130274d4b1e06e2fd1e6a936a9bcb156996e17e82b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-143.pt","sha256":"ae449f74fd50272b8ad10a2ee60521d06f3abc5c9ac96d82cf58288453ed248a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-143.pt","sha256":"f6da5a05b49d5a521873653ad3216029a12d900637a0b345fb08b3fe7636a4f8","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"2ac42a24f679b708982287ca848654a06ec67a4f82b56d3c68eb0c865af8ea8d","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b38cafabd16cf8321a81eb37d669a14db06d964aa68e49834a7797228514ba44","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0b86a5f6d8887085fc19a165db95b23f154d99f32e80d2f44a2ba97a063f037b","loader.blocks":2391040}}},"awc8-wsd-h120--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":120,"attempt":1,"job_id":"2482622","recipe_identity":"d8a416a40fcda3f5941119ba37864b1b5c91bd70344855dac9a2d9a5a1b09589","loss":3.357420486152484,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"98a0e98fab7f168d3ddd74d9959c3321730c4ee5c4b1b9168a26d55417b717f8","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"8736407dbc97f894d5734dc90ab00bf3d3259abe064d3536375a74b22ca95af5","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"9b93348d18697bba7a2d0bf03e7de919ff09d40e4f90b9ac52307a457ac46665","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"b0f452075aa0b5ca79119ea4296d5973996845edbcab369753039d2cad57dffd","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"cdc942bd7e5a9edebb358842a669ab0be4a161c3bc28ca295b61d67fc4ccc9cb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"72903ace2242e8ebfc073dbee9de2cde6db2cdce81a432be1f1bfe658eb13e7b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"fe82ba34645bc12929708567f73e171655fa72ba93281630be29a8ba4615c9dc","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"afd0daf4d34e72e5338057be4cb73461400b8ce3f1154d433f3572faac8e7cb0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"71205e829f98d17518b4dd5210c99bef7765c92b914bf4ca1472566e5b20aa9e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"0b495c470234db0a90d71780f8abe91dc2da8e4200e79732479a0eb6b304aa6a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"c4d8f9e875350b42c656c11118eb484e542ae529ee97628e37b9a911a033cac7","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"84ebccfd27b44aafa8507e0b66f59f150095616c003decfd6bf7cae1277172b9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"067ee25d80dce401c084b3d04b306e38ebf10d413dbeda11536962649f9a81bb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"76767bcf8e38b0a617798d0f0ea32fdc08b206f3fedd48787473a0f31f496dd0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"266305d54c6bc6906dcc12be2d012cca9281055ff9273fbf1635c85f9f42022f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"b09ce243e78acc277f888d71bbbf1b64a0a7dd5dcd23cb6575e344aac9fc1167","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"961004658121086e78f84dc0353e411992a7c4d27ffaa7b222a85c1e2d21572b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"ae0a193f02cdf4aaad41abb753f2d8656d6953604378225da88b1c0e8b76b3a7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"8edc5eb652abfa9544578c52a1358f23430a2587a4713a70e742d8bc673ecf25","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"7399f44b0a5b6ffe130f8babb6ffc10df3f6b03af16d0c354fa300cf822c3f50","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"c36ed0f07aef24b23032e56ad073d153c146a81c073f8343b588ff02e87a568d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"661ae6123a5e831679d006343a8246086d10810d957db086ee4ffc573f1cc2df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"4134eda656693b6bd159359197a7b99a58864123164a0ba6aae4ced6ee083663","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"96fa2d82fe6684448f630081103b9960219b0b560d25831a5ba58e423b936367","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"0358e9f67881e0f9accebb8f156117b1aba32f1b5a7a5852c7e0fb5479282bb8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"e9bbc9cc2546c723092908a2584366e9e1f1fb504f35479d1f83f386cad064ec","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"7043ab42e02d101bfdb137c9bb952de79ee579a22d15dccdd8fa24f60f7ccb1d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"5fcb7b2ec11fae66ad4f9e39036eba349a6854b9b0a4bee91c04e459340e0a0a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"2f3693a357b5669ee2b39263ef6ae56d034ed5a76bf8672d0f7706919921bbe8","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"89804daa9dc414911c135b68b7c1e62024f421328f0d32a0d7b4f5a77ccaafe6","bytes":5343}},"requestSha256":"ac35a1d32f9801900c676c7890f5bed5b8f34848547c21778eda33c1d3afe647","metricsSha256":"d32bab0b15c0d1d9ce17f482a07041e2feb5c82c850398193ecabddb6ff7043e","continuationSha256":"48b9585cb42f8812ff685559a34f850b604bc57f0c9019b7da763c5c32dac671","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-143.pt","sha256":"c36ed0f07aef24b23032e56ad073d153c146a81c073f8343b588ff02e87a568d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-143.pt","sha256":"661ae6123a5e831679d006343a8246086d10810d957db086ee4ffc573f1cc2df","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-143.pt","sha256":"4134eda656693b6bd159359197a7b99a58864123164a0ba6aae4ced6ee083663","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-143.pt","sha256":"96fa2d82fe6684448f630081103b9960219b0b560d25831a5ba58e423b936367","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-143.pt","sha256":"0358e9f67881e0f9accebb8f156117b1aba32f1b5a7a5852c7e0fb5479282bb8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-143.pt","sha256":"e9bbc9cc2546c723092908a2584366e9e1f1fb504f35479d1f83f386cad064ec","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-143.pt","sha256":"7043ab42e02d101bfdb137c9bb952de79ee579a22d15dccdd8fa24f60f7ccb1d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-143.pt","sha256":"5fcb7b2ec11fae66ad4f9e39036eba349a6854b9b0a4bee91c04e459340e0a0a","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-143.pt","sha256":"2f3693a357b5669ee2b39263ef6ae56d034ed5a76bf8672d0f7706919921bbe8","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"abbaa232c793cd0bc964ee54764832d9a4ce859b5cd1e677d805715762c44f07","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"03cab47858b55e02e969f32ba79b9b866e2a4ffab84e8ee29e5784a82f3df72a","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d8a416a40fcda3f5941119ba37864b1b5c91bd70344855dac9a2d9a5a1b09589","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":120,"attempt":1,"job_id":"2482623","recipe_identity":"b66b61f02e00729b78d54de3e5614281e0401911f630f4c0d280d701c3e4f956","loss":3.370138168708808,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"3b3a6f060313fb22ad66c7803c260398fa64c168dc79ba435c7d328dab116636","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"c3283f2d22e998fb3c06b6dce17915dd98d2d08688a5bc212e7a17cfa97c263f","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"162cc98364393e3d0cffe79e0f0577f548250199ed2229eede5421103b24fe2d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"c6f8f8e78dd5707c8de24c2a99082e509c660475b374fd2089329b37c2311915","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"3a682a286fb76a759cb0ff2130a8d14fe614bbfd8dbcc15d5cc7da56982ad0e2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"a2a52fafdb7251a98bc74ac1f311f22539249abd9c0cff911f89e9fab811239a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"f95a7accacc5f29c475c4bc4a8743518d714a2c1b4abd700d92089741c058c36","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"d4b75d9e703f4b5c8ee37e33bb32325813c5ff62190cf7a3df9992214e81ce14","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"5b80222c525d9ed6ffb5f67868866f8a838756eaaae9178d27e6cc41d904fda4","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"7055957d19338b43fdb7456d77235e51fe11b154534d8efcddb1fa476734a7fc","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"4b6e54961353b72bb4883aa365c513ec221c922018614b0850fd0f9c6ef4c331","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"a0d8231e33e2fc09daf90bef695709e7b758da50b0bf643de348f8bae04e1d9b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"8b73148c05b2e544e4abaf5cc3e9e46fa29f4ec5f26cf78be18872c3af55c51e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"d917286c9dbee3bdfceea821e82024120fea79ab8c24205f99863279b7dd250c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"454797fdf04d111638140de9bf3f3f3b6f7e64555b2650fb35645197afd9e1b0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"3d97728a3e62b8fc222d959606f416d94d2a898c3e824cdd42fae953fa3e1fbc","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"9e4c240e0a3463892ede3b8142e488ddcb46108e4cc226541f8bacb35aa4b4ca","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"9cebc6fc98b7707275bd5b358d91426a7912e4ad67572a3396a005c06aa83683","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"3b4dccd4c989d6a093979fc6de41fb3aac529723afea2e1d02f26ad0e2b0888d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"9f7accd715cd2ab549d47a0e0b96fd7be9a8a4e777c80b36f220a8cdc1a015be","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c348217d0d02845b469c21993bf745e2693bc49b78ac1fc06b91223264c2fbf3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"c91e2fd958936618746c692d71fd2333c52ffcbdff1c09b071e18129c06e571e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"91d37f3cd4483de0cb8f272f42a57d0df5b45e4e4763866c5b760099447a70ff","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"6df6c499694295de032d4bc2387985e9f0887f6cc0671a52f661269827d0b5ce","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"bc1fb668d17594c926e9458c38beaad3d41023e7eb4d79bcb42ae94fe5ddd62b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"e80bb44d911c3ee031334dccc351b6479df2cccd54e8040618ef721b81a32e6c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"36f7eddb8375d408be05aea6d205431aabd0bcac02e3eeefb2467b09cbafc286","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"2f115b5be10ac40fde3be6094e234833f3ea4c3815431cb17f74c79e4c7676bd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"56808e711519a1205bf19c320045d8c38fbab238bb86a5bc4f6ff251545d97ea","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"b2446b73abd679a01aab3f86d2e8785f300bb433992857c5e85b87528b836c04","bytes":5257}},"requestSha256":"d9b37059353880bd9079ffb2346613f5325b8336cb4c4648696be233d6755738","metricsSha256":"56b415fdaec9de5a0e82fb0fdb0d8533c9b6c0b00cf8cd5b56423b6624685978","continuationSha256":"93e2b67f24405985fc77f1ccbdc1ed683b38083f8fc82bbb192f2667613fea94","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-143.pt","sha256":"c348217d0d02845b469c21993bf745e2693bc49b78ac1fc06b91223264c2fbf3","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-143.pt","sha256":"c91e2fd958936618746c692d71fd2333c52ffcbdff1c09b071e18129c06e571e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-143.pt","sha256":"91d37f3cd4483de0cb8f272f42a57d0df5b45e4e4763866c5b760099447a70ff","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-143.pt","sha256":"6df6c499694295de032d4bc2387985e9f0887f6cc0671a52f661269827d0b5ce","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-143.pt","sha256":"bc1fb668d17594c926e9458c38beaad3d41023e7eb4d79bcb42ae94fe5ddd62b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-143.pt","sha256":"e80bb44d911c3ee031334dccc351b6479df2cccd54e8040618ef721b81a32e6c","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-143.pt","sha256":"36f7eddb8375d408be05aea6d205431aabd0bcac02e3eeefb2467b09cbafc286","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-143.pt","sha256":"2f115b5be10ac40fde3be6094e234833f3ea4c3815431cb17f74c79e4c7676bd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-143.pt","sha256":"56808e711519a1205bf19c320045d8c38fbab238bb86a5bc4f6ff251545d97ea","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"762295856f327d4510bea0ebc8b5ab7771be9794932eca02ab6f11402db88abf","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"703838c8661c7739dda29096ec11c1ab9ac0b9463c57398b830e6a5ef54531af","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b66b61f02e00729b78d54de3e5614281e0401911f630f4c0d280d701c3e4f956","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":120,"attempt":1,"job_id":"2482624","recipe_identity":"9c2f048e5f5b28c44e993287fbdd7bf42efd10dbd25d932c0eccb60e5454f1bb","loss":3.3736219692634295,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"e6510cf8bb1a591c490439ad4240afc769464601e2539219b40b50fda20d3daa","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"707c685cf763666921ad83029d6372bb736635a7d389bf5ce6c70e3aecb84c99","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"ec1c7fe9563a809420b7223e5a291e953aec56199c819d4725fb2ba5e41add08","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"737ae0fb58927ba15ff6350ad5503e9c569be8d2ac414e4115a79f5df7308595","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"f68a4271ec08b9eb93a8180f3fcef4a924ee75c95bc75f476912b827d825217b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"0c7cf3ff1b3b48fa260e91ddabbcfa7b2e7a37abe4636f70f1c99714af6be292","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"e7f7a5d5cd09cb8e8cdcad4eed106b8631aafc9108e80465624601fb3506bfd9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"e619489509e64903757f6769076947b772a1becd2c794c150ba023a6f04d8c86","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"a29317b8648e0e732ab985620b683284960fb795495a9b9841097fd3e05153e1","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"1cb7f04bc8c390749ac87abeaf67be08148d174e1b5f01357bf444a8b357ab7e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"9fead4ffc19cd4d171840cce7fa9515cfd0a6df4fc0b3e021cfc5a3930a9680b","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"9fe148fe88e88f484d6646d5a8714d30ea44c97ad581b92571f8f3aee63181bf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"35caaea33f67e8b7f0ee02594ab6418b5cf523f1d1f62d9bb527adf8970049f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"ee452db3d4f91359eaeaaa0cf054b524fc9b70c6212a1d9a0b977daf1b0fc728","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"1d9daee43b2b59a4238725de54f3edf3263b74b3a966ecab63084ff17001f1f5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"f73a2c9fee445c13110d6c71ee525ea83134109e0ae79919542c1bd445372e2f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"cb099f44ffcb8627d9a5c0faba4702779f8ac84fbfe188ddaff8ff4bc78f4e82","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"91f3f486e18995d159596eefbec412cd884cb10339e137ee32ff20f1b98ef968","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"40984f1f037912c51f209bf9f43625896e4359b4111290c58fd162cddcf28ab5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"6f5c35ce2febc184c4221c90d3df130bc46466251cef8627c5b9b570e5f22c8c","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"678f11fb48e68e240a7e07c24185857ed25de4d345dcaf83b6bfe02925c144e5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"e08dd502f93353f7a9350d80acf8d90a7b37027d505fae0b280cb86f01fb49e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"2e5908220c432eb6ae964fcd71577bfcecdd3b04893786795823118290a0436e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"c686124b10d6fcb2a8fa56ec74aaf669e279819c4ef5692c528c27f445463248","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"8370d408864673eab0e9e9d8e2e8c0f5646eec2a9f82e96f562bbb40c597c77f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"7259b058f58444697cb6d61fc2a543e560d47d0fd46bcf470dfae4f6742e8e2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"d9107fbcf460f0d93770f10f360ae3f54dae96116621db3267e6d32662cd38a2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"3e2ca745e0555142c0a6be786713b9e22118d0c37d541a186667669fd0c3f4f1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"acc9d33d01e0cd2e839c211082cbeca972a8c6fe528170631ca6b1d4d1ad2dce","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"b086986a6c50d99310f7fd753efba61d44eda73cd752d29a55bcdca876ed78bb","bytes":5343}},"requestSha256":"f9eafee53c931c174f655a8806afa359ac1b4cfc75d41d44aa7eff74e05ac00b","metricsSha256":"f37a4d583c242c5db68bd0daad72ee5a702010ba05b05b5b2f1522c5cc7f2d62","continuationSha256":"49e174dc5ea1777056d90033f9b5277a22695845a9cdeb99c507b875b3e22b9a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-143.pt","sha256":"678f11fb48e68e240a7e07c24185857ed25de4d345dcaf83b6bfe02925c144e5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-143.pt","sha256":"e08dd502f93353f7a9350d80acf8d90a7b37027d505fae0b280cb86f01fb49e0","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-143.pt","sha256":"2e5908220c432eb6ae964fcd71577bfcecdd3b04893786795823118290a0436e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-143.pt","sha256":"c686124b10d6fcb2a8fa56ec74aaf669e279819c4ef5692c528c27f445463248","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-143.pt","sha256":"8370d408864673eab0e9e9d8e2e8c0f5646eec2a9f82e96f562bbb40c597c77f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-143.pt","sha256":"7259b058f58444697cb6d61fc2a543e560d47d0fd46bcf470dfae4f6742e8e2f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-143.pt","sha256":"d9107fbcf460f0d93770f10f360ae3f54dae96116621db3267e6d32662cd38a2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-143.pt","sha256":"3e2ca745e0555142c0a6be786713b9e22118d0c37d541a186667669fd0c3f4f1","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-143.pt","sha256":"acc9d33d01e0cd2e839c211082cbeca972a8c6fe528170631ca6b1d4d1ad2dce","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"9bbee444d269d161a0ca35b075d55d7cdc2412b3bccea6f0633612eddd282e80","bytes":1958918037},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"97ebdce4b9b90a0b5a45a1dfb2ea127a43c8b64f1e01c41c53aa46551053811f","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9c2f048e5f5b28c44e993287fbdd7bf42efd10dbd25d932c0eccb60e5454f1bb","loader.blocks":2391040}}},"awc8-wsd-h120--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":120,"attempt":1,"job_id":"2482625","recipe_identity":"e27c90e1e90959565f648d41149532628af523873b645d237789a82bba9aa15f","loss":3.3688452132349234,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"043a0bef4856727ec06ce3eb64c1b39109a77740e4fe7e7bf2286470f30faa96","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"f655097640adbb886e1fa780ffcb98776e2f41ac181e2c4aebb8614761678132","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"b5bef29cb5dc645d82fc1f16a6619d59cefa584af343cc36f9d56f5bfed7016e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"f6be0c5cea1227c5d38f548d5ff780d167a3863e930f61af18167ecd6ff7fbda","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"55778606f424b40cf283de0ea90319d1ae16d01d6b55262c6160b51154a67b04","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"6bdd3875de398223877b7af18a9f425a7621b46ab21748144a8255b2c0ab4be0","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"c244832765eb6b1d6ffa25522b5e3bccb8119c292617467b6f8e9697431148fa","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"0e31bad599bf988dbdaf274a876cefa3353890ba351f968f7417b439c4f3eab1","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"93d8a3761db0b39feab0d089f4e5eb50ab281fe12fb74b9287e2214476176094","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"42c52b369b1d12e2931df659765c8c138d7edb0e65b9e826c3767d8ac822b55b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"170469d70a510da1e3501a5a6a0ff8f124755fa1078fed259a7be345d9aeb5e1","bytes":81621488}]},"pre_checkpoint":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"b1fba519a038bc6f02ac844ab614f170bc8c849898a81921f6fd0ecbef53c9d9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"ceb426ab8196e0e7f9703eac54f40dcb08ea015af161e2440f04094d860447db","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"d3d67c21569f7bd18552ce9d89ca252a3a8f0c612b8749c90a5c311597c74292","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"8375c9964076fda4f573d9535dee2b9c0a93a2bb2b54c73879bb98dd17c1bb80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"bdc5e91f1b74b61d4303ea7f7a9ddb04e44493bda61c331f0bb9ad51fbf5fb42","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"24eaa0263439edb8a9fcee72bf22bc5c820f36dd00cb4009b541150ba6707a7e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"a9011712e16b9a34e8d405626e3b68b0d5b38430e7c547d5df66f98df37a407a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"97a4784f7837868ba0daaed00dff1025581537299429adbb9869bb3321020993","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"c2e8a0ca8cfb4008de1a209eedb7a557e974ce1c556e10ad0f7ba83d6330df6d","bytes":244926383}]},"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"f0529862cfaff35a04380f7e72b9a5db6c4680ae84088ed04c28ec4ca38cba4a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"b2eca46b0b2c10038dc88ab82f417bde9ae208086fca8a039d0f1816f343b77b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"0fdc44c6ca8c9ef9c5920ca68935736cd926bd0eaecf5efe1821a333ecdea0f4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"d91b4c22ba58fa8147bc4c48943d85a1c8243d5e1a23a99eb591c278d3d0690d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"0c2b9449e94bd895ad2b0045b9da1c20157a54312d903dfa54d0863cd459564b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"4567eba83df13d7b25283d45c1d5822140ef2712e7a7215caf817b1467982bc9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"61fb4ab91ab05e77cd9da2d8b899f7a048c18bfb9a99161448f4103c2cf74228","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"6a87785700f929b8e27e9ad8f887bed86620b22e3cce6fad37941e84ccacb372","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"d27012f653e2abbbb14615772a74a7ebeeeab9dfe603d7cda292b1b4e23fde80","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"dd252505858454e47edadfef562d8beb57644dd5544c3c8f64dde13aad1d6ca7","bytes":5315}},"requestSha256":"be34dfeef4da0e38ad263ab0301f8af079b536f35e4d808d815f498ce295eddd","metricsSha256":"4316b4a26dba635297901843f4f1acf3713fd231ef26c4d2be59f57bfb835cd6","continuationSha256":"8864557b8d2f14e49a1cd7d8c154a6a5111599a4f247e87b6af547aedcf7f9f3","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-143.pt","sha256":"f0529862cfaff35a04380f7e72b9a5db6c4680ae84088ed04c28ec4ca38cba4a","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-143.pt","sha256":"b2eca46b0b2c10038dc88ab82f417bde9ae208086fca8a039d0f1816f343b77b","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-143.pt","sha256":"0fdc44c6ca8c9ef9c5920ca68935736cd926bd0eaecf5efe1821a333ecdea0f4","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-143.pt","sha256":"d91b4c22ba58fa8147bc4c48943d85a1c8243d5e1a23a99eb591c278d3d0690d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-143.pt","sha256":"0c2b9449e94bd895ad2b0045b9da1c20157a54312d903dfa54d0863cd459564b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-143.pt","sha256":"4567eba83df13d7b25283d45c1d5822140ef2712e7a7215caf817b1467982bc9","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-143.pt","sha256":"61fb4ab91ab05e77cd9da2d8b899f7a048c18bfb9a99161448f4103c2cf74228","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-143.pt","sha256":"6a87785700f929b8e27e9ad8f887bed86620b22e3cce6fad37941e84ccacb372","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-143.pt","sha256":"d27012f653e2abbbb14615772a74a7ebeeeab9dfe603d7cda292b1b4e23fde80","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"884af0a1482fcf886838ec40995237d724320185bbcafe0e60f4a6a7a7168402","bytes":1958918101},"source_horizon":80.0,"target_horizon":120.0,"epoch":143,"step":11130,"cursor":1424640,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":80.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[143],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-80/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"590a0070fd123dcc19744fb2ad01a50aa8cca1f92488804ece0544bc50bdfbe5","loader.blocks":1593984},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e27c90e1e90959565f648d41149532628af523873b645d237789a82bba9aa15f","loader.blocks":2391040}}},"awc8-wsd-h160--0.003--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2487090","recipe_identity":"fd627886ba06a1601ecc7e3ac52d91fba700acc099b2a4b06923fbbd14988169","loss":3.5063614500926086,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"b95ddc893cc39d1ab5d713fabcd726afa77a168e7bdb8b0ee835fe841146326a","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"de2d6762cc4289f810b3ed92a470ac9ced860a3affd0196b22146cd3f6bda6b3","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"ae52cf1eb308996f2db04eb45112cd8fd51caa21820a97916b6928bdfab3b10c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"0955d418f036d848536ab76b7806c6627d10eab9ffb46cd788c16b50f0fd1361","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"3052e8419e0c876f24d83d6eb2391dc1cac306590c82b42ab36f1dc5b06d81fc","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"eb2de9b373cb39246d63153f26ca53de3e463385823bc9b32380e51e8d973716","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"9f41ce2fdedd539605f7fcbc750acdb798dc58acbebfebc0a66124836dd9dac4","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"30a30585f3aebb19ae3625c1042022eabe29d2c3ba343cacd6f455c511659baa","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"838d3bbd0d37355e390457c99516760af7e1160d2d2262296ea5f9dfc1a37b13","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"a680fb4f6bb54bf4266d6118c1034897746a85174e0570d92c8a29edc4370187","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"a1852440b76c0c1f147c1dea255367a05f5ec4c758f806a4d530a18cc5bb1c45","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"79deebc99fbdb8d1100a5bdeda8be721689f9f8129e3c55200d91a3850e80680","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"1b0760e0c330272aa089d18b4ebd1455e5e7b5582936e5bc89a97f0f82a9d34a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"3063179051286258922332543de8b9f43435509dfa53cef73086441007b0e4c2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"1528b8314b24d76acad9ce9f332593315aef99eb691b002bf65966da23f6917f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"09a3d7a7b3726387a88ae5ba55850b66b63ad1e5881faf02188df706d9e6734a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"991953ca74fae39300db5d1f078130d134c868a9f6a2ad6c175c5fe760abdcd4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"dfa6f50851c4137b616ab8be2adae138e69fbe9f52b4bed82cdea61317f95c7b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"a04d0e6b0f99c87e86b304f117dab798a33f3a6ace5651d66cc8ca3edc2c6819","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"5e46943d2166070205be4969096bc9ea708ebe09268ddfb022279baab8358520","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"b6c93624682f6a4b5b1c50ccbd82f67a8181f03df80d5f956a6fd29cbcd34976","bytes":5341}},"requestSha256":"1ef0be1d009d5c3e207da590a1e49acdf3cd001f2306528fa3975e383e1a374a","metricsSha256":"84232c8df3769d30e646594e8f0f4fe4ae866b0034b84b330dc29fb913c7dceb","continuationSha256":"16be5c50026b063a814643d0be6b486f709b7847cf51fa1e521a82f624f5556a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"79deebc99fbdb8d1100a5bdeda8be721689f9f8129e3c55200d91a3850e80680","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"1b0760e0c330272aa089d18b4ebd1455e5e7b5582936e5bc89a97f0f82a9d34a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"3063179051286258922332543de8b9f43435509dfa53cef73086441007b0e4c2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"1528b8314b24d76acad9ce9f332593315aef99eb691b002bf65966da23f6917f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"09a3d7a7b3726387a88ae5ba55850b66b63ad1e5881faf02188df706d9e6734a","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"991953ca74fae39300db5d1f078130d134c868a9f6a2ad6c175c5fe760abdcd4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"dfa6f50851c4137b616ab8be2adae138e69fbe9f52b4bed82cdea61317f95c7b","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"a04d0e6b0f99c87e86b304f117dab798a33f3a6ace5651d66cc8ca3edc2c6819","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"5e46943d2166070205be4969096bc9ea708ebe09268ddfb022279baab8358520","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"e8b6cf08c939cb69007b134966e009a65ff21532630949bda9d28ea4d0bbde85","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e14d9486565665d2ea54939cb9db23cbf3e05b7c089029d6d31a61b2009a07fe","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fd627886ba06a1601ecc7e3ac52d91fba700acc099b2a4b06923fbbd14988169","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2487091","recipe_identity":"a54649c9e11b2eda4dc185a6bdf4fd1a6dd713e1c20956d6989673dd7f089b7c","loss":3.5025594780223552,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"e1ae58bb2ac9281d8f99a4e9f17a98d98f0fca9315548812783a6c316b6388f2","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"43f4fe4f31b7335d6b59b001ff68e40d2755afa9729478794225fc59a1ec6b7b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"1c3561e0ead587a92f933407280469b3cddc0a5778413e856d01bdda9bb5d707","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"aa73f5deaa5580414a5c2f2d54fa3d4a90ad319b5201028144d29b925217aad8","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"9d59e8409bd18d9f089594dc3ee938a2c34b08d229b80020d595252907355dc2","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"45fe4d9aa3662b772d6a6d361f2d7781d6b8775bd8a447783f876c21fe267143","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"63cc4dd9da0a5b57c4d2e36c872aad8b05148916ab4f53f04c3417890dd60c2b","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"7205a7c6e67e06be99394230c79a4808af77cab5cf26deb9659d4db7a67393f7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"25cb43fb0da958a9c5d1f2934b0893bb2ba9d3055e07b500330a2ec0ed1caa3a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"a376fd3a2da47bfa5ed6f857f329f159382a95ff0d0d66dc8c0a6e65553cc53b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"5f5ec2e643848759697f148e12c6737ce13f9320fa95e72d4e1784db361e11c9","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"897f28263a4aca0d7fa0f9870733601363bb41bc4ac7053d0303d25a4ceced11","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"b619f4c16427befc7f332ff7b365930fbb05ca37620a6e6a80cc86cef28ad5e3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"c51649f13e08a87c2219da25c7a52eb47ae81383087255ab3646defb1ed0ab31","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"83a35511499b29397db0956d460b62d9d09b263c3a8177af080d25c91a859ef0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"84e02033cf27b77cdba6cdfb4234c3496721304da30694b7021810447a7a5d73","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"7e76ce12930fc3621723136bf50a91cd310a47f8c7ae1a3ff6204ab42b181748","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"6a69d26bc91c09bb1b53b3d31749001c2191924f219236f7403865d2f665b834","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"eb67d58d7544ca25b892ae2e1438fc881c85eccc9a8a947d95cd126531f2491b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"323eb44f26ee1950fa81bed0e4bb5086c35517259d2ef6e3afe2653eb81cf576","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"07679c9a9d05514ab01e7f1b6eaed51c0bbf98e3edaae637ff14bc9a1c127070","bytes":5312}},"requestSha256":"0eb73ffada1ec26b011e2407e83fec67171e1173df666be95cd4087a87fbb022","metricsSha256":"b33027809509948f1c00f9f653b3d514ba7c229a2b03cf36298a2ef9dc15d3e4","continuationSha256":"428e90305c39f2de6072959352eafc41815a3bb600a0edb8b4ea05b47a254ff7","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"897f28263a4aca0d7fa0f9870733601363bb41bc4ac7053d0303d25a4ceced11","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"b619f4c16427befc7f332ff7b365930fbb05ca37620a6e6a80cc86cef28ad5e3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"c51649f13e08a87c2219da25c7a52eb47ae81383087255ab3646defb1ed0ab31","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"83a35511499b29397db0956d460b62d9d09b263c3a8177af080d25c91a859ef0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"84e02033cf27b77cdba6cdfb4234c3496721304da30694b7021810447a7a5d73","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"7e76ce12930fc3621723136bf50a91cd310a47f8c7ae1a3ff6204ab42b181748","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"6a69d26bc91c09bb1b53b3d31749001c2191924f219236f7403865d2f665b834","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"eb67d58d7544ca25b892ae2e1438fc881c85eccc9a8a947d95cd126531f2491b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"323eb44f26ee1950fa81bed0e4bb5086c35517259d2ef6e3afe2653eb81cf576","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"5dd9d64d4bff7f2df6057487df10f712d39a2f8b0906b8a4689f7cb255c7e741","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b7a902ca136f3ec5d27cc19f53ebab0cda0a26f7c173112f61551c1ba30e5b9b","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a54649c9e11b2eda4dc185a6bdf4fd1a6dd713e1c20956d6989673dd7f089b7c","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2487092","recipe_identity":"19aef0661a52f446bbbad2a0eaff2457715126d8e5561dd81904a6a0b2d9c7f8","loss":3.5105157180067326,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"07b9cafbc066fa4a701bc0659552ce1c9124a2adc45bb78743d7b1426f51068c","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"4e13bc157aa1ec7bccaa5ee782708205c750547a29da4dfbfafe1f3e4c43b082","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"2c69bd61bb1a9b07dae7aadb39e64d7d40f1f31fb14a4ba0fff886417456a358","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"d2d1a9adb429e8f01518bd10592ba90178fe87fbda7c27ba9037e89acf1f026c","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"70f57cb5890c4d7a0921cc55ca71a32236a2ff7fbbe2e49dccf452de8127a2d1","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"2c1c9d26e2d16aaf1b5b88259aec53929cf5316e2709bde24dd6834506fe64ac","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"38e08db7c274e2992ac9a44a00f62d080980859a0398712d9b397b3e75313003","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"5d7f9a393a397e3d51cb8ea74b52dc8f5e388b42dbf29b4723d03bc017dd3045","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"ed3ab9c841cb072779e0a91998567f38326d74eb748eea26a7eed6ae6afaa77a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"8091cd5b2f2ca675b7c1ba90d7bb0a17d85ffa60210f5241af40a33c8b3f5104","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"32c35142c0add5c38844759cab54ad60378c7fcff025e661c9e59eebbadb11aa","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"fb0cc617a4c251094dabdc3a194875d3b20081273dc3c91940974a762fdfbb97","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"10870c365e2a0d22a01ad80ea819c588c7b3e7291be876205c8a92f9bffa91f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"cc476cddf4e393d756be7e2cfb20f9074658d90a520329975669cae502f51613","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"7e7c2ed1a50dfe4f01d00c10d3cd01e9705f1f48d14b90891391642c4308d99b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"31c6ab03885e091417b44b0f79dd23e22484f9324cb419d4ab34e970d24cf852","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"c0c6d7c383a0eed78aed3b25c7041ed61ea50c6fd12176f120a0eb35168a0822","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"a2bc20578a679db2fc265ae901e43c762a75861a158e0439c1915a6fe2df1264","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"84405428e11130475bd209b98ca01326529ca4afa6df150d2523956f2c1d1135","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"409659315c8bf016e260e6bcc620a326cd605cf8e3161f4d393f7b7f82e990af","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"9730463e1fe8e12f59efc4603410f78347bc4ab8b16584e254e81dbb81f85f7a","bytes":5256}},"requestSha256":"d0ccfa32a3d3b3067c54d6af3ce6d8d860c3161926b41902ab650ab490fbf60a","metricsSha256":"7150a3be925babd3c4272e6db19b1df36ea3834f0e59c69cf39bcc4f559df731","continuationSha256":"9b063eb397e7e0b8ea76410ae87e714fb2e7228ba73075e3166978b1cfce42ad","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"fb0cc617a4c251094dabdc3a194875d3b20081273dc3c91940974a762fdfbb97","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"10870c365e2a0d22a01ad80ea819c588c7b3e7291be876205c8a92f9bffa91f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"cc476cddf4e393d756be7e2cfb20f9074658d90a520329975669cae502f51613","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"7e7c2ed1a50dfe4f01d00c10d3cd01e9705f1f48d14b90891391642c4308d99b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"31c6ab03885e091417b44b0f79dd23e22484f9324cb419d4ab34e970d24cf852","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"c0c6d7c383a0eed78aed3b25c7041ed61ea50c6fd12176f120a0eb35168a0822","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"a2bc20578a679db2fc265ae901e43c762a75861a158e0439c1915a6fe2df1264","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"84405428e11130475bd209b98ca01326529ca4afa6df150d2523956f2c1d1135","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"409659315c8bf016e260e6bcc620a326cd605cf8e3161f4d393f7b7f82e990af","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"6314616edf10fcd84711b1fd5c7b999bb8b870d2f16213d2f7dd3f718145d8a0","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8ebf89405196743b4d09cf2bba223566fa8ae360cd2447f1b45d37eb31fcb2aa","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"19aef0661a52f446bbbad2a0eaff2457715126d8e5561dd81904a6a0b2d9c7f8","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2487093","recipe_identity":"9d17a9807435af56a4079a23762d50486db2723bf3c305fe14a1a52812780860","loss":3.514450208243573,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"3a0ab5fdb323afcceb0c95074dbe384773f9947e830b40ca4e36ce7e7b8f0bf0","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"543d622beb90e445652e177dca4bf2a3a9026fc775ee83afeaa996b36b8a2fba","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"861ae335068d518058aa0898adaf17cbf2c16806fbbbff49ac768c511a474ef3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"c138c3becfa1424453110ce26b9a86deb1f40b3213be4aaaa4b465db64ebe6dd","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"2f5332469ae5f2dfd131d873a7ad1a3239eac538df5af0fb33d7ebc807b0992e","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"4ed0662f6b814a9c1255d0c679e0f67965d918fd78dc7287b6213e2967f2c9d7","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"e4b003a79a378b211d3d8c11c83f51ec596687547465a6d2b06f97c84a323aec","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"f10a6f6f798f84f44ae97959367adaed19a8ae5fdca6c9735a8dce429843af1c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"f86f7f5176f269c1436385fb16dbc01a3efc04048320b69ec6e35d9abfe0f2ce","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"7f9c7d57d66cb5355cf23ce9c1e1b517512f45b7ea70341bc52d5d980df11af7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"ef5224d8d2bd7c21f48bcee4b9eecbeb02999081c9e3f13837909466e347a87f","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"dd07b436aaff0c11bc21dba45ff06ab8e927d4b904e7acbbb7679bdcdce1abe9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"39d570a4c24ecef81d3f33cbc2996e182a458738d97ffc7c99e81dadcc770461","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"161a6be4dab2500aff55f9216c6298be3b8ca9896fdc96d66427e1753ea03665","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"8110862e1e1c039d3c8405abd8ed78c9652b88a90606e526fb9973c59b159b26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"d080d9dfa4a40a03665c1dab7d1bab0b17485d63358e686d42cd893986a88d1e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"b7713938873f68d2325f67a807141a36a646cb61cfdc680544058b8dfb9dee90","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"c6c555ee4c42256e635601ecf4acd787ba12b3200ae58168a2f2144e8ca5ebc6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"099caeadb630402bab7629d482acb28f4ba3d9136137cfff78e52120a0b05e97","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"f1b19aa7159e91392620d6f3b86529675acae14909d41409afeabc68afdaec3c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"0ed483521b4f47e802c4b27f6792d4684c50ed6fdcf0de1f24f6ed46801e6fe9","bytes":5344}},"requestSha256":"ce707090489929d758b5c5423ab7b479aefb45e8b67134faf6bd2138a59bbb62","metricsSha256":"30326fb3cd547a9c4e6f56a6cc22e0ce3d7b8e780bf30c285c409b3ffa830fe7","continuationSha256":"cafd09e918f9fe88c8707289a431ec9750360c478c0f5eb038309a10acaad399","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"dd07b436aaff0c11bc21dba45ff06ab8e927d4b904e7acbbb7679bdcdce1abe9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"39d570a4c24ecef81d3f33cbc2996e182a458738d97ffc7c99e81dadcc770461","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"161a6be4dab2500aff55f9216c6298be3b8ca9896fdc96d66427e1753ea03665","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"8110862e1e1c039d3c8405abd8ed78c9652b88a90606e526fb9973c59b159b26","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"d080d9dfa4a40a03665c1dab7d1bab0b17485d63358e686d42cd893986a88d1e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"b7713938873f68d2325f67a807141a36a646cb61cfdc680544058b8dfb9dee90","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"c6c555ee4c42256e635601ecf4acd787ba12b3200ae58168a2f2144e8ca5ebc6","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"099caeadb630402bab7629d482acb28f4ba3d9136137cfff78e52120a0b05e97","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"f1b19aa7159e91392620d6f3b86529675acae14909d41409afeabc68afdaec3c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"02aabcf70e2a6bdb440a1977da4d5621ab98036b39f9af418e92289d87131755","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"21f1491540b88f787c4189e1e3696c9b13e6d281a771ed884926fc595670357d","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9d17a9807435af56a4079a23762d50486db2723bf3c305fe14a1a52812780860","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2487094","recipe_identity":"2130dd8b1b23cdfcf1802fb432b9f78425d36f77864d2837b124192305b03ab1","loss":3.5094643790826194,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"82a80f940c18ffa4391887feb5205b026b10ef0ccd73fecd4a0dce9fd7d2c70a","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"1d4fae9e2b20fde107f4c4e8b5b79a0e4d156577516edca97ec0dc1b0a777bb0","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"7e3d304a94f6f278eeb3e377215c9a668107c47414623c9f77b7e6dbafa83c37","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"0c7e0c8a53cbfe31d1cc485cf7362fd6110aa95c826cc5b60dc031843528857e","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"8f4341cc9586149c31aae5faebc56ceaa71f96a0811431103a3d4ff2bb465b45","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"8592c6c336edcad93eb2aa057fffbfd5621d6034b52818a7fcdc7039b5ce940a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"aa284746c1b6e2808bf37cceb1ad0e597908885c8ff4dcc4e371c596ed414497","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"93182c73eed2c83ea9e1a041de25db3925d6bf88009bb15d3e80a4b108302288","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"6ddff556a305f31468dfb32a9c85f475ad0d8187571644d99a8d613facf65369","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"e78929b132f29e73204e2ec2287653050ba4430a56ce6723b77d7767b499ed53","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"8c390c6865e94e7de2381896c825c79bfb29f9906cf387f99d7f126537d0d64e","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"a937fb53980bbf4ea33b7d4e1a9577f933edee493df07156b2c2d40ead7272dc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"9be1107a417a96cbeb88eac5bcada0df4bab328d6b83b2972edb064c8729dafa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"152e14e89c7904eb0dac627d97832db159a043ff26661bd8e1a0a4ea5c5756ce","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"48e8d253c52e4b0b316e5c8718bbd28be83dff15860e9b42297c6a1f0c471a9d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"75e502b70e5e09cf18c1a77465ce415876a1939ae4155e04fb53cdca87a58604","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"952eebc4989e2311c523f829e90e668264ed3d63660ad6a4ef4f1bf88f5d0c5b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"0686a869f28cbea83c0f880c61db5eb98404cda23ebe36b068386e086252a784","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"15c8a8a05233c69475d102145f72e91fddfc1b2d497ec639e24e6a8d526f800f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"d412cc4114ee13b95e39ee4af1c0da6a4e4451d288ce8f3175d2b12087ab3632","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"57eed638981df8ccc95c220f216f69867e46a48815af7138e2aff61c869dfa60","bytes":5344}},"requestSha256":"4fca979c1b8bd988659f3d3dfef1b6a56009085a9413ba751fc05fba94aaab2c","metricsSha256":"6b566afb256318684ca79cee79390f6112dcebad271bf1ecff07f1131a51f60d","continuationSha256":"34a109ff983ade69c6496790f53f630743c60105d28684998d14db9ef75b5dd4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"a937fb53980bbf4ea33b7d4e1a9577f933edee493df07156b2c2d40ead7272dc","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"9be1107a417a96cbeb88eac5bcada0df4bab328d6b83b2972edb064c8729dafa","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"152e14e89c7904eb0dac627d97832db159a043ff26661bd8e1a0a4ea5c5756ce","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"48e8d253c52e4b0b316e5c8718bbd28be83dff15860e9b42297c6a1f0c471a9d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"75e502b70e5e09cf18c1a77465ce415876a1939ae4155e04fb53cdca87a58604","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"952eebc4989e2311c523f829e90e668264ed3d63660ad6a4ef4f1bf88f5d0c5b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"0686a869f28cbea83c0f880c61db5eb98404cda23ebe36b068386e086252a784","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"15c8a8a05233c69475d102145f72e91fddfc1b2d497ec639e24e6a8d526f800f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"d412cc4114ee13b95e39ee4af1c0da6a4e4451d288ce8f3175d2b12087ab3632","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"3b89ce2c7fee7fb47726b4286b99fd045472343d844e83391ad71eb98587871a","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ecdceb353dee6714840128e40f5e4f54fa6e974af82c66fb2c5ac2223214f1d6","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2130dd8b1b23cdfcf1802fb432b9f78425d36f77864d2837b124192305b03ab1","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.9-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2487095","recipe_identity":"414a4abca0354aa2d24eb6545d5e938cbd0ff645db331a79814c3c2b353dbc73","loss":3.517926365905667,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"d554ab9396436672782aec7fb77a90861cb5376a4e998d84c07d17f470811ec5","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"4e3313e5667f2aa59d6c022f71fae6eb564fa8680735f5723ec9e14f0dac9dd1","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"2f87bdff67eda458aacc2b75a6a4e2caaa0df3f893558ae4679307642cde4f86","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"ba693958ef2cd6a65770e9b595d7ac9eb1b8ad1f86c2a6cf117bee1fd6168251","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"7cf7e5970f0eb8309c63e13b75ee2d30e01983520ec33bf31b0e6f85483c6bbe","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"f02e9af9ba8bbaee91a780a523cde66f9e90abd7bbbf46d74deab5ce74fd336c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"b7403eb63074ba15d3f0298f0dd25802833a0446a658191ea31adeb5b5ae02f2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"d44ea97c44b9a79f35971bb0bb2682fcb3f85230d68b88d65e27290de51394de","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"dfc68a5c0f741f2c21a12771c6662264f7865bdb51634b0ed0eae1b21707a2eb","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"1a5f9219e7c4a386f45bd36d5e4f4c708bc56b2df9440b697e5e9be1e34d8958","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"07a166d5c48319fdccf0d0e4b4bfb008cd74827d8d04cfdbf3767f47e0764a1f","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"a437f0ddd35813d7fe33c5133968af05620ea38217364fb56815781bd5729f7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"8763601341437fe8e01fccf8ae1ff0dbedee7698537bed0b416803be50646d02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"54401e5ec64f26fe0fc7440af95a7315797b2665c0f77ac26c034e91599b2f1e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"5a2c4bfc6e559369f91d35949a0a80a5c758f9dee4ae7d31b7e493ec8dd76e1d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"16fe379c1a072681db46113f7e8ea61e8b1d5975ef3202fa8efd9772275e0462","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"6cdca0594e2005d48f7759970036dbcc9927e61aedf28531b2801d96faa65fcc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"2ef1da29bd6207e50f52305579965dd56d41e520d09cf0d007b35a47045b32ff","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"77092a37177297ec7feb8a1d169c75be672fa522f948cfaf4c67ee79ee248523","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"8d1d15b299c5c2ec8409f151d68b984c846d72b7e4482137944e05fdb988ad83","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"7eb37a0a501252c2a0460205db016ed66441e29693f55d8c6df009bd66d5cff0","bytes":5258}},"requestSha256":"4aa4fec743f3079d069350dd91e3a220100e411288fe473ee497966a5f406c40","metricsSha256":"500f8ed60ad24b18eb00684bbb3c1fd3205101b6a6708c03950ea33c310c802c","continuationSha256":"646a4ecb5ab979157d8ddccf68bdff8e0a057c59606d43e009c15d499fbf13cb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"a437f0ddd35813d7fe33c5133968af05620ea38217364fb56815781bd5729f7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"8763601341437fe8e01fccf8ae1ff0dbedee7698537bed0b416803be50646d02","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"54401e5ec64f26fe0fc7440af95a7315797b2665c0f77ac26c034e91599b2f1e","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"5a2c4bfc6e559369f91d35949a0a80a5c758f9dee4ae7d31b7e493ec8dd76e1d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"16fe379c1a072681db46113f7e8ea61e8b1d5975ef3202fa8efd9772275e0462","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"6cdca0594e2005d48f7759970036dbcc9927e61aedf28531b2801d96faa65fcc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"2ef1da29bd6207e50f52305579965dd56d41e520d09cf0d007b35a47045b32ff","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"77092a37177297ec7feb8a1d169c75be672fa522f948cfaf4c67ee79ee248523","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"8d1d15b299c5c2ec8409f151d68b984c846d72b7e4482137944e05fdb988ad83","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"7e1a8ffb96cc9b73be1de09f73fe0d529c50ce71c3f1254e29b77e6df1f4ea2d","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5fc7942e76a196423554e6b438618b4f9ba58b246580b99d2cafbd3d5ec31f4c","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"414a4abca0354aa2d24eb6545d5e938cbd0ff645db331a79814c3c2b353dbc73","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2487096","recipe_identity":"0c863ed6ff17bdc587881077b64b8266c58f51ebb279c27fcfef429410b98f13","loss":3.3969474862469444,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"ad299e8448a849756d9239841e3ef139e8abbba771c94a481060a2f0aa8b6688","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"42cfa412b7f8d94c8fdce6daa02a420e74c75e5bd62fe458f8c03d6673e23b2c","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"d96cab803c645b25c29e46389f06fc45bc0bd75f396ed0cb099f6104623ed960","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"2e1dbca8396c568ae94c0ebabadc6cdacd8d9db02172000071aff430950e5bb5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"ee3ec8408ea44762d6c62612ed5301e2fdb4bc976ff7eccfe7bb4d8b8dae2b85","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"91130b8938cd8fa5f094a74b48f46d5ff5a2710d7e6cc142a0c381cbb3bcd9e8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"be4e9b59f38ec298c902d69a1da277dd0a6452cf742c765f6276abd3cbc4f009","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"148f4217525bf08acc89e0389dff7b64bf9ca7bd5a3f7fb8da4a5be1bf77341e","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"d3bc48e6c031be7a5b5ba17f74356c860584536522d25e7c3993826915a9a787","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"c2134e42122b89733f27bc5ccc94e415b3339a92b5fb975ea07230f4811ece0b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"f854eac97f1071f15ced3ddf7cf9941e8afc69d49c3c6c3891eb4982ca1b7294","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"223f19b3e2318725b7f1af30b2c313d26855f99ed9afa3992071e6db27d4cea2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"1efef8666699c0fd75b8346dbb8647ad67d0eb2e305a8c0c976afea22d221f20","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"c4aa52db8e3937a3c91f2bc0e4d04f68c08271dfce47be3e4dadd657ae0cd8cc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"a972981341aaeddd5ca25c2fdc1f18f82801195fb9d997377b511fca78ea1ec0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"4524f3b2dde442407caf340d37402a0b873f057d87c6acbafba01f1efa83a1c9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"41c0537c5789338edda9073327882a2cd21bc3e3f24405b2cd30ed9093dfebdc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"ee37090395388f9ce46a0d9180418a0685025aa9cdc86fc052e047a3f2786b67","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"253f7f8e946a678c25cbf7bdf0f461b67185a8b9a8da33f6cf0d6389c5f7a64f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"ba0cc4f4584f7478f95d1645aa08d434949219cb80f95936a7dc8bb292d874b9","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"bd044d744993115f0fadc23e3ce8ba25abbaabb6f056bd699a1d39c89aa2193b","bytes":5344}},"requestSha256":"0517e20d49396a5f1a098290911df9f90acd4f393e0b31578b8b8d658b688a2a","metricsSha256":"8cb5140fe76435c1f30a8d136eb09aeae812df3bb93b4075e39d2fa67b5dd99b","continuationSha256":"3fe376050494b2125dd7acc129d1861bc1c1742f0fd2b088774a68e8616cfa78","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"223f19b3e2318725b7f1af30b2c313d26855f99ed9afa3992071e6db27d4cea2","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"1efef8666699c0fd75b8346dbb8647ad67d0eb2e305a8c0c976afea22d221f20","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"c4aa52db8e3937a3c91f2bc0e4d04f68c08271dfce47be3e4dadd657ae0cd8cc","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"a972981341aaeddd5ca25c2fdc1f18f82801195fb9d997377b511fca78ea1ec0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"4524f3b2dde442407caf340d37402a0b873f057d87c6acbafba01f1efa83a1c9","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"41c0537c5789338edda9073327882a2cd21bc3e3f24405b2cd30ed9093dfebdc","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"ee37090395388f9ce46a0d9180418a0685025aa9cdc86fc052e047a3f2786b67","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"253f7f8e946a678c25cbf7bdf0f461b67185a8b9a8da33f6cf0d6389c5f7a64f","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"ba0cc4f4584f7478f95d1645aa08d434949219cb80f95936a7dc8bb292d874b9","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"14e1045687b1e21485b1931859f70a1559b8d994adc22c9164b7c4db3f495a34","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"95a69f33712f12107949a8e3f873b9a7e15ba4b4050b4c67d166f4930e0add42","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0c863ed6ff17bdc587881077b64b8266c58f51ebb279c27fcfef429410b98f13","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2487099","recipe_identity":"d6308fe6700cac259d2a9178fdf3635e2161efebf1266279702945bc2f364d0a","loss":3.402039755179342,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"fc878c0e51dab96b98902fa611d307589a178c2f0193c3984914beba33d8e823","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"2c86060c9fbeb35391632147f6d4285df86176ab3c297610a1720a8203da794b","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"095045ebdf24648f5c4f23ad7ed92bd702a36c69b524fe5ab4f6d7f5ca5e2cc2","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"47d3994c3d5dced77136513710f608d43f4d0fa95bfea189981a77d7fd444dc3","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"01ad80a108f46fdcce52deb3e09cf262ec2bc630d9d1c3a81245caa3098a30e0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"fc3342bedff25bb24d13de821b1f6f090207f79adbcc48ee1ed35af57fc846d5","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"2302c79cf0a89c817405eb9f9a418d763a57284811590996d9ea6b9b250d988d","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"69eda233811f1b5cfe1e169dd35e3abd83a08e8966c4fc8548cf7e7f91bbb945","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"dc35519050aa4445a738baf29df66af0629f75cc412707cd319a8f137ac9c54e","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"f61f15756aff7c1566c798da193d7a81f6bc64542b4fbbb9e8097a8ea5cc7ab1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"c2f22327e24b616311d34d2ce28b0f6f7fade140ae92552e2c6c1f966a837394","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"5fbe452069d93b1948deccb13b4049911165130db85d9a0470e0489dad455950","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"d9ee9b166d3b692ce81ac828fa2c9caac3bf664e1615eb4c103b9997d4f0ce6c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"f2dec92ae5e392439e27b672d05383cd7ee8e83855d1b5ff4234c95b3ad0139b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"4a759b2bdd04c2c8421682d7b982bf5034c932af58c8480c49333994f1bc76a0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"a14bf1b6d8a7e3455cf9d3847d48718ff4cdca053af4fa86479d1a73f642e115","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"0e8c54dc3a3b0e386b4abc3ef3b29cf3aa58db21ab5e70f4d30b08f95bb0e89e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"b5137b539965a6a8474d1cd0a90712715eea8ba735a6ea67e85f3a4442a7b5fd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"3fc3b9113ee6ab9175fb95dffc5928181583d4f0ae020de51f0a0c937e816bd8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"ef5f9023fdd546701385b50bc19914ab71bc3e47f071c1fb536dfbe99f8c19ae","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"3f86c2078611f78438d249a97066cf9680a27d89cadc8af5ab19a1b36527a047","bytes":5258}},"requestSha256":"506f868287ae5368b12a874f306cde9e86b9cd3807dea9ea7922130feb10c96b","metricsSha256":"2076609a5109ab32f4d8297ff4d9ff991dd58eef9ef0ffe9dd26c101fb96c0cf","continuationSha256":"b0ac3f1bb66b84d6b0a4992b444fb5f96734c21f39a28aecd070d541bdfee7a0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"5fbe452069d93b1948deccb13b4049911165130db85d9a0470e0489dad455950","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"d9ee9b166d3b692ce81ac828fa2c9caac3bf664e1615eb4c103b9997d4f0ce6c","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"f2dec92ae5e392439e27b672d05383cd7ee8e83855d1b5ff4234c95b3ad0139b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"4a759b2bdd04c2c8421682d7b982bf5034c932af58c8480c49333994f1bc76a0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"a14bf1b6d8a7e3455cf9d3847d48718ff4cdca053af4fa86479d1a73f642e115","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"0e8c54dc3a3b0e386b4abc3ef3b29cf3aa58db21ab5e70f4d30b08f95bb0e89e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"b5137b539965a6a8474d1cd0a90712715eea8ba735a6ea67e85f3a4442a7b5fd","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"3fc3b9113ee6ab9175fb95dffc5928181583d4f0ae020de51f0a0c937e816bd8","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"ef5f9023fdd546701385b50bc19914ab71bc3e47f071c1fb536dfbe99f8c19ae","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"87bd008b030740d24d6dabb60ecb43b1e5581b8d8e7bc9b9a929636e53b5a8df","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ec11577a5cc4f74475bffa54661304f8c8d9d21c4567a0001809463d8d892e39","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d6308fe6700cac259d2a9178fdf3635e2161efebf1266279702945bc2f364d0a","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2487100","recipe_identity":"e3cab499e7daa6d1c6e86ba8a5b86f04e8b462598477a3e78ef7b648a7a14226","loss":3.392649371669298,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"72d6dd922fb6d1e0cd824e4d780279374185da516eee7e054c377321f770d690","bytes":885},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"2695a4c3876d945ca8eae9fb38a6c022b7c4df6245150b1678a2e17c825fe3a6","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"b981a6d092de2542b08ac1a2a9425945cbbb20ac9d0874be490141b89d1db8ac","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"8ad42a401fbef4ed0091cc235aa35ada5e84c337eb82a8ef2c03bb74f18fb2f5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"51487844025f7097bbfd7441d5cc92b39a957c12e5e420f78a198dd5a27105ed","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"ca9a8a7b353c590dbb2b275156093d130f2cdb3fc2b9f0a1f73546a7eb494473","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"ddc20af9dce4fcfc15e7aaeaad8fb4b6c98436ab4af091d892ec966dcac85918","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"94b7e4a1296682037e05302172b6c3b9dca40a94572ff71a307663e0a476de33","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"2f8c923ef2466599b31c75bfa21be456eda57acb83aded0167e51200f1d059c7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"121c8429092049b5736f67ec000b052ffed38cf2d465619e20564ca878d4be00","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"f718c6fec8a5212ab6f89f49baeda898455ac3398d7b3618be65492bd12852b0","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"633b7e11695be37f8289f1ba94df4c0a3b650a1d701584be4d0aa98266a35954","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"080e43eae6a77925b0e5659b517a7eae7159696afacd56cc8e5d09a30d108373","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"dd2052506078b5889ea481ecc4411d34a3edd4381d9d5abc1431b5ffd8763033","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"9ba2c8cd899551de5fb3e9852496ff1283e34ee62833518c85d1a6237a1de49e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"c95bd14a1101b77d2d6a9ce0c332775f7597b03deb0b2e90c7c6ea7094447bdd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"c3b1a4a1d793e415d0b0b508305979feb46d6fd4c23a58b1654c9cb7b6e68a76","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"e29c17585e68eb4b13963a9542dae6202afbbaa901c2ca9b0dba7132d74cc183","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"0bc4a88912b4c36b26af38416bd0c76eceaf8a01196d3b82673c01ad625c5031","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"b78004e475cf6a3e23ddb7978220abcc176c757b8e6ffa201e39bac843e7390c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"e8639a717988d35d4863c1386d3638ad82ed893968d72c301bbd31a9b98a69aa","bytes":5344}},"requestSha256":"9c4d9a484db78c2bc3dee64796ab4efc293222feaab42d1e50fa81d2e0e8d767","metricsSha256":"169301c2af12197f21ba5432499e3bd10044fd52845a13520568c48d22a17302","continuationSha256":"0c9331bedb067b3cf91d3dc0fd7fb1c84c0ff02050fd705c9246933d21d6971f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"633b7e11695be37f8289f1ba94df4c0a3b650a1d701584be4d0aa98266a35954","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"080e43eae6a77925b0e5659b517a7eae7159696afacd56cc8e5d09a30d108373","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"dd2052506078b5889ea481ecc4411d34a3edd4381d9d5abc1431b5ffd8763033","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"9ba2c8cd899551de5fb3e9852496ff1283e34ee62833518c85d1a6237a1de49e","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"c95bd14a1101b77d2d6a9ce0c332775f7597b03deb0b2e90c7c6ea7094447bdd","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"c3b1a4a1d793e415d0b0b508305979feb46d6fd4c23a58b1654c9cb7b6e68a76","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"e29c17585e68eb4b13963a9542dae6202afbbaa901c2ca9b0dba7132d74cc183","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"0bc4a88912b4c36b26af38416bd0c76eceaf8a01196d3b82673c01ad625c5031","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"b78004e475cf6a3e23ddb7978220abcc176c757b8e6ffa201e39bac843e7390c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"25f6ec858ede991489de7e465371d6ab2e55f3214c34a4571c32a420c469e263","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b8c41d9534f8bce0bdda4614fa428ab5cdb9dc03b25dc4399eeb70aedc252dd7","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e3cab499e7daa6d1c6e86ba8a5b86f04e8b462598477a3e78ef7b648a7a14226","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2487102","recipe_identity":"17cb948e796a9a3ce5180526073581c635984b3adb2dd292ffc98df3b6d8eb59","loss":3.40863865097916,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"3c407034143e927939097daa95796c6fd4674557460ff7fd58f464d5695ab422","bytes":882},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"32f2db109e3e8c99264c11c608927a8fcd6101b1e3b9bef6e92e7677ee406938","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"cd93ad5776d89fc6f573678c00dd220bb53e18fb1d4794453beed3aec0f11379","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"70f0447443bd09997e2847c4d5114994dcbbe31b721e04a5f465b685e4e0f4b6","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"91aaa6a019ccbdcda45983601db514e6a86a978c1305477742aa402ae7d35d16","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"f5752e5b1faa484793649be777ce544b7451b53c4e89b8ffcc1b8e04f69efa9e","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"fdc3836ae77ad191936541b9389c1e6012b9366d609a09e1ea4014373cdcd499","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"602112a54596da0082b0707944e867cb9e4c2afe19030a9411e01e1063065c9b","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"2e56e9ccd22cb49513310e8786f8d65d25bb1d8c7aea8c58692fa3763a449e36","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"31d7e1deb6f7c43f3202dfd3c715f79f47a31e7ec60dd6e7018a663bb10313d7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"62081d647ebf985a26b0728baaf563133cb61ae12fcc59eecd78b141c36aa5d2","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"d032e6a881a8d9c917410efb53ef35d996d10945e1d44aacbaa46565a7c962e0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"2e855d4bf2b738e00abd66f2864451f6e53ef851043d16070db42ee55f30e8cc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"a46128e26cb29bdfbab762fb3516d595f77a620957fffd0b9a54349d14fde0e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"a0205b03250cb7152f1905b1359e3554750b96d8c5121dd7606fa33b5b154c55","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"d08d8a5ea5645e536b10d1ca94afc848bb7834a008ed62ad227bbfb9ce134ed5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"2dc5780bda3521f1ced2d565123ff0b435267f85965b63819c9f286902703a10","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"53cfd3234088d003c435b4c13250c2b64b9f08e00b099cb12010e9b5b1fc0b3d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"6047ffe97775fb5903bb984423f9b0a346315fc386ae0be1f6ed9d6eecc61c32","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"649e62cbd29338dd597bd4fef33521f676dd522621a778441212c2ea091bf9d5","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"ffc98eae00fef03b368f05b0a6461b8364a0ef774217bc803bb3cb2689a5c4b1","bytes":5258}},"requestSha256":"088addd8b2dc1a85830e504c22a6c785c80d82a7f9779acf306cdb267b3501c6","metricsSha256":"d2885480efb7095b0d7c19fb1f410d6106daae25c06a7720cfc364e9ce6393f9","continuationSha256":"47823016e777b077e6916cc2c9aa1b2887be3d49ff264428629dedbab8a18fd1","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"d032e6a881a8d9c917410efb53ef35d996d10945e1d44aacbaa46565a7c962e0","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"2e855d4bf2b738e00abd66f2864451f6e53ef851043d16070db42ee55f30e8cc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"a46128e26cb29bdfbab762fb3516d595f77a620957fffd0b9a54349d14fde0e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"a0205b03250cb7152f1905b1359e3554750b96d8c5121dd7606fa33b5b154c55","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"d08d8a5ea5645e536b10d1ca94afc848bb7834a008ed62ad227bbfb9ce134ed5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"2dc5780bda3521f1ced2d565123ff0b435267f85965b63819c9f286902703a10","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"53cfd3234088d003c435b4c13250c2b64b9f08e00b099cb12010e9b5b1fc0b3d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"6047ffe97775fb5903bb984423f9b0a346315fc386ae0be1f6ed9d6eecc61c32","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"649e62cbd29338dd597bd4fef33521f676dd522621a778441212c2ea091bf9d5","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"0cd1da6fa0364179a6c5bc2a493347c99a7ce8696ae4841dc1fa55c33daa78b5","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4c0567ca492a2d7b60d1c8fbcb75abeab397537c2f8a26a15a5019da7c33da5e","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"17cb948e796a9a3ce5180526073581c635984b3adb2dd292ffc98df3b6d8eb59","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2487103","recipe_identity":"5e9e85153b210842bba444ce79657b104640dda9cc007b19c9406722629eef15","loss":3.398285878117938,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"86b9e54d6b957b1a293bdd4b55c0a9eeefdcc3b9e98ed04bb8649b0354fe868d","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"c4137b86105d7ff02b7b8ce158dfc08f3d67f3199de3cc070d7dbb28175b1a72","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"d20214b2ed5dd1ece817cc9a7bbaec5aa7dc51a66a5e793fdbf163dcc1a26ead","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"0586bb2ca80c94a1b51c478aa3d69040afbc0e7728c98c3aa37d138990025f57","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"6da1b1ad8b570d8f7727114cadfa322651d278ae28e02ca2bacf5915d5b1bc02","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"f2e2181f0137b6f0461cf1b900cfccbe8f766c4be64187a8e79108b274528604","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"95c017cde92fed8afc45c4f5d565c11338a88fd187a480ca7f996bbbe19bab03","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"d0fb2b3208db7db9807910c076174e4a3e52e722f3817f819aeea2db9f85c8b5","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"5333926a7f10db5840c7463fd5137952eb81dfb2f7ee4509747baef8dfc708e3","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"3a4f507e22f5fec577c03abf9a3c4cd080ce0af9b251c550f4ec4b412b5b85d1","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"50678a2f67827d682c63a31f381840de501df6da8cd144710cbb003081e0ba58","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"f5f130a598b047fab53c94e1643b1dff54e0c5897608ddb8741ec943b6adec83","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"fc892a6bbaef19d66bec6e49d3baa2dff6d23a1065cc3805955c86894339b062","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"b484658a0984b382f8d6027e182766a7ca9c4a80218fbfc7acc2597944809460","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"026ee95ade59b25d214fc9dfc9b187e36ef2042ea734ffc85b60619cba85688a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"98ed3958f0affb2ce61b10808c34905ba0728cae13c1a2918a6f579b68d532d7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"e76e43583fefe44c8090081bf00bf4f3fa135ac7d0ae1703d3fb83341e64a7e2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"57c4413fd08e71777dbf47c3591dbe70ee9e32d32117336b7be132fbb0a011b2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"c6c98a3a9183e9c51dfe1ef3e2d400543cc69a08e86e5d5fc1b72b5221a891ec","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"faed6b3a46039c06ce6a69bfdf5f0a0a733a939dd37e273dbc17f7f82e4f1822","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"c5d75477fd2b466e54eb5a2a43993403ffb3d369b937be98458bed72e63c9a44","bytes":5310}},"requestSha256":"e49f08571dd294dd0fd2eff3af0cb1e58c3cdf984fa87ad1171e939422c1e664","metricsSha256":"c06f08093a46499dd65b665348b2427209b56ca08e8c0874057b42b2ec822aea","continuationSha256":"aaec42fb8314eb547f34e174068a74dae8337cb9f037baf5910df454ea149a5c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"f5f130a598b047fab53c94e1643b1dff54e0c5897608ddb8741ec943b6adec83","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"fc892a6bbaef19d66bec6e49d3baa2dff6d23a1065cc3805955c86894339b062","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"b484658a0984b382f8d6027e182766a7ca9c4a80218fbfc7acc2597944809460","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"026ee95ade59b25d214fc9dfc9b187e36ef2042ea734ffc85b60619cba85688a","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"98ed3958f0affb2ce61b10808c34905ba0728cae13c1a2918a6f579b68d532d7","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"e76e43583fefe44c8090081bf00bf4f3fa135ac7d0ae1703d3fb83341e64a7e2","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"57c4413fd08e71777dbf47c3591dbe70ee9e32d32117336b7be132fbb0a011b2","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"c6c98a3a9183e9c51dfe1ef3e2d400543cc69a08e86e5d5fc1b72b5221a891ec","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"faed6b3a46039c06ce6a69bfdf5f0a0a733a939dd37e273dbc17f7f82e4f1822","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"6fcad9200d4ddbd312848b530e907dfaa3d7eb627b5f677f30a1618312291bf9","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8d483290942b186734c93c48692b178961bca881fd92251c6f3e15e44bf4de55","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"5e9e85153b210842bba444ce79657b104640dda9cc007b19c9406722629eef15","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.98-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2487104","recipe_identity":"ad851acce0e3f9d9d43c22bf0f564a48db93c0058171561333848609d72a1974","loss":3.4035677666090716,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"9a8c7a722be1e6be49f70360d843d628434ccf7642a64e7e6fabe7a4ad24fc42","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"bb02366b1a925ab5dd9663dee2491f4c32c1c64931e098c88d9b452349a497fa","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"47762338e5c6ec801e16119f0454987194e83e86a9dd8b9e65e3a48f5de43dfb","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"60d1826846cb92b194d1915222116fd668fc412441d71d7dc59b18cb0b85dfb2","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"c40cb340fd3b04a96aafec3ff4cab7d97fc3c4bd9532215e8aa58458931c2cf4","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"8455397d33f5c10e8c9667ae1fcfb36fe59407b58f37f45af969de91cedeb1fd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"057e9170a251667adedcd3b99ea4c7b734ec08bfe3cb86c303d33a50cbcb79f6","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"67f96dafe80f313bd503ed2b2392a3c6c2b3c069b0cb752674855ad98617e360","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"0bab37d6212c65d031debb5b359a54baddb0ed23c475d45c4072ab83cf199062","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"b14fdd7de9f594a94ad3639b03479d60c3106d2efe708bfaf5405be0386fe736","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"578fa8c5d62ebad79d6eadcea142042ac298101a3cae5692aaf97901c1d6763d","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"fc1f9cee24636715c1b2f5fa945202b260f0fd89369c6b58aa2cf45eabf34e7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"f405da431c5eeecf2d7069fcb01170698a3d74c20b5f97cb69a27ad15e7b1a17","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"0c9650b709b8428e6584ef704887707910cb556f71ad8808f52ee874452e1989","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"76ccdd9cd11627844acbd1f5abf2535ca6507c228e3b6deec5dbcdcced4999ea","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"5749bc2a170a2785571f31bfa5096a0f672fd4bf9cb73b166683692399e37b91","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"a1543867623d89a4fd8df640384811ebe6b4e86b6d44683f62f8998fbf8e1a67","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"a373855e2103ed11fee3547aabcd74ce2cc5a3ba86536a6f5a54d50338626410","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"864c556e3345995233656a6fc4d7385c317dc6abb0e7b9f67bacc9d3ff53a795","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"617ef33fa8b558c6a0c92966fff8fdfa5150661ad746becfb1106a93ed9b44ea","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"214bf56172a47e2e767a575b4a67f5afa0c2f62511a473dfd39e52c2b6911194","bytes":5256}},"requestSha256":"dc9d352a195bd0b54277b4c306281e59438cd5550d6606500fd58521b3b69538","metricsSha256":"2adfcd2f2191cd1763ee20f575888814af3080eb8261bca2eac4a79df03a5dda","continuationSha256":"4df7a940cc04df4c713a26ac0d76dc6422e7779ef6ebdd4587ffb30fc70cca81","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"fc1f9cee24636715c1b2f5fa945202b260f0fd89369c6b58aa2cf45eabf34e7b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"f405da431c5eeecf2d7069fcb01170698a3d74c20b5f97cb69a27ad15e7b1a17","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"0c9650b709b8428e6584ef704887707910cb556f71ad8808f52ee874452e1989","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"76ccdd9cd11627844acbd1f5abf2535ca6507c228e3b6deec5dbcdcced4999ea","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"5749bc2a170a2785571f31bfa5096a0f672fd4bf9cb73b166683692399e37b91","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"a1543867623d89a4fd8df640384811ebe6b4e86b6d44683f62f8998fbf8e1a67","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"a373855e2103ed11fee3547aabcd74ce2cc5a3ba86536a6f5a54d50338626410","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"864c556e3345995233656a6fc4d7385c317dc6abb0e7b9f67bacc9d3ff53a795","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"617ef33fa8b558c6a0c92966fff8fdfa5150661ad746becfb1106a93ed9b44ea","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"5e938eba9626bc7db15b79050fcf51d24a59abd49f050ae886d9bd2839f84335","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e27ed699c1199f0e866a57bec7f05eb05c665d880748ff2d824569fa6645af04","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"ad851acce0e3f9d9d43c22bf0f564a48db93c0058171561333848609d72a1974","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2487105","recipe_identity":"728b18dfafd014f17788eaca69736b6d33b69f7d14e140c646b5c291f9369604","loss":3.3529225996304977,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"e7a230f279f35a70c0a626edb7f402643f870f4c6ef0a6549221ed6542abce87","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"adf29f1814c328cab88108af6c8e2c195d3a75102151fe70ab5e03656f90eeab","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"2f560b5fa948b7243f979a60f04ecd09175a0b9e8d1040bf65186f6213fa5db0","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"536083b29304af84c52456154531fbed94e30abe70c1016e876e65e4294d7675","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"0ecd53e290f08442d67e02b3f6c9d981a3eed5320168546047ecaab042847ee6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"e37971dc9efc5a269eb94c9eac3a4b7e669545772fe9c37e9d2440214cdca0cf","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"c16790f362377404687e313550f7f1059e20861018543733bd3353a14f4268b9","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"a2eba3befb510dbe2638c3028eea18088c0908655853d2cd99dbbfd543420fa7","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"c0fe754cf91bc50df300f3358018683a78c6c29f82f92b2f2fbaf232854431dd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"69b8a49a01ed0be4368807f1687e58c3d44353b1c45c55e8f419369180a975d0","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"bb662aa026809c9cd7f679ff5fdb52a77415c2b704cb7a2dd8bb79294ce56087","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"7058ee955955cc38232782421355f938a61b7382de5ea48b9a487b4e610712cb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"db0baca5f56edd6d2a239e71845053b9b5258787c1f3b42645552ebb71e8441a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"5fad8cec5a4a14c7e94a6296e8ce4208ebc99a2ab1ef419a35a0ab059ecf6555","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"2125eb767af82733c4572c3a3cda8478438222bef312dac58eca5cc76a6f50d3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"49c8051c2239c3ab75dffa2f7986fbb0fafaa6c68c5271c6e8678c7f7d5dad7f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"22a732f47217215b46c4a89dfb4543901a102e3015fd89d77ec03d8f33049d54","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"4fb65c55b463e28f52288d06c239855400d06b9d17548160f5c5ce0f29670ab0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"ff189d31e41e67782c652c27bdff8c5e670926d2ed23426a51ea808c0d5a8f87","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"472ffb7726ca3cdfce20263e14a9628520d22dd12b2451ab35873ad9c825ddec","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"36ca01f904a472ce8c4707587a7e7f194a1e09cab5324c2e4b617f7dab150e96","bytes":5258}},"requestSha256":"0e73ee4dde002b2bdb850cbd63a551dfc374526e788e63dbb2d8608af74fb671","metricsSha256":"25e6576a6b9701239a1405dfbcc146aecb8c9f76fa30f7841b7ea5ef1a160dee","continuationSha256":"7cfbbd8537ffb3f0f7a12a330f353e2d55e1e7c094b0fe14bf270e88434ab05c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"7058ee955955cc38232782421355f938a61b7382de5ea48b9a487b4e610712cb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"db0baca5f56edd6d2a239e71845053b9b5258787c1f3b42645552ebb71e8441a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"5fad8cec5a4a14c7e94a6296e8ce4208ebc99a2ab1ef419a35a0ab059ecf6555","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"2125eb767af82733c4572c3a3cda8478438222bef312dac58eca5cc76a6f50d3","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"49c8051c2239c3ab75dffa2f7986fbb0fafaa6c68c5271c6e8678c7f7d5dad7f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"22a732f47217215b46c4a89dfb4543901a102e3015fd89d77ec03d8f33049d54","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"4fb65c55b463e28f52288d06c239855400d06b9d17548160f5c5ce0f29670ab0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"ff189d31e41e67782c652c27bdff8c5e670926d2ed23426a51ea808c0d5a8f87","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"472ffb7726ca3cdfce20263e14a9628520d22dd12b2451ab35873ad9c825ddec","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"fb2a045d481bb3f260f32e7068ae5a953a4bab7172729c4d6bd082959c4b3311","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"068ff7ed9340c26aa55ec2db14309c51afb30dac74875affd32bc26110133bc4","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"728b18dfafd014f17788eaca69736b6d33b69f7d14e140c646b5c291f9369604","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2487106","recipe_identity":"dbee3b1c109b4476264a6965e0c375a334b3d1fdb78b03bd3790eacdf2469ecb","loss":3.359074351485768,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"2f2ef22abce0c46d2bed06ea277fddc487fb07cafbaab2d45bf378f45ac82d88","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"0479893dff768c56daa547e3871e7a0f01314efcae908d61079c1091f9f588d0","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"641cc46cda3562906a6df32de225c9b18ca8520364c393b8211230b00b5deccc","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"99de1b9ba3d3c5f44c1f3cbc603924c05d4d727f1049148b389f0264641f2b2b","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"57a6567df174651851405faebbda6f82a6df42c2d007216ff8138a0b07e28c8c","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"a0dd80632c96215021368d73ac6016347a131f292c060c7935b1f9825c2b7819","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"8bdf5b5c8fad580e21c19755659d70a7ca0767bd5e657815aa70c45845a712ec","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"bcc580b1f1ae8b670958dec1a1051b626e48e82301961ed977309b8d991ec5ca","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"3f9817cc2644ae24da22eb63fd4e4fb6ea7c36672f7e6486deb5e8704be16486","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"b835af3fc81f17e1e1cf6e3690871b95d4a0b30eebb68b6700448cc0cb77f3ca","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"7c621966994ac19bfe74e36d92476aa67393f65c7739bcdc1768012c961725f0","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d38d74bc58ebec5a25b3d6fb1a7a7cccf8bca544bb959684820cff0182738e41","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"c689576dbd91576b6fcdea4b0a5bbf3e26c3dcaba9a0abbc304b920bab2c414e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"c5c08666e1d682babde7fcdda8af89eb2cd7cb5704c903299a28b0fccf3a4f72","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"e3493b45856e1844caf4aa795fb4fee64595ec0ba6ffd661237d45a673ce1611","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"c5248b56591ad3383f2b333dc75361e641701e1055a9455316442c7c4807a10c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"5a3d004aaaf771abc062293a993c1c9c51422aa601a68b323c122f5246beb4f7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"3ebb4a8941ff7c88bf86c646c9e37c559a0d82befe92c2eb8c7eb974cd231af3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"287890962691c0029e42efb9a78daa0cba27131223fde6fce9d9f6222cf57ead","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"af7e7b34aa2b994ea9133ddddf8a7c4e8b3c7e23d4e032760227e949c6abaee6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"9974fe370e60388a0bb334dbc10ab6f2dddef02251fd5b51a18ce42d5e7375c0","bytes":5313}},"requestSha256":"011e2cabe2ead1e6400045b46a84c04cf76c52c1ea3cae9db45e72041498e9ba","metricsSha256":"6f924d26e6c7c2a849cce85a80b450d29a347cec2e9b95bf59d78295716c81fd","continuationSha256":"b5d9839d22ab0a693165205cc1f2b3b197c59db6f6d44b3d98763dd8c1888b05","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"d38d74bc58ebec5a25b3d6fb1a7a7cccf8bca544bb959684820cff0182738e41","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"c689576dbd91576b6fcdea4b0a5bbf3e26c3dcaba9a0abbc304b920bab2c414e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"c5c08666e1d682babde7fcdda8af89eb2cd7cb5704c903299a28b0fccf3a4f72","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"e3493b45856e1844caf4aa795fb4fee64595ec0ba6ffd661237d45a673ce1611","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"c5248b56591ad3383f2b333dc75361e641701e1055a9455316442c7c4807a10c","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"5a3d004aaaf771abc062293a993c1c9c51422aa601a68b323c122f5246beb4f7","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"3ebb4a8941ff7c88bf86c646c9e37c559a0d82befe92c2eb8c7eb974cd231af3","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"287890962691c0029e42efb9a78daa0cba27131223fde6fce9d9f6222cf57ead","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"af7e7b34aa2b994ea9133ddddf8a7c4e8b3c7e23d4e032760227e949c6abaee6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"5feb295210eeab40e42b7cf6b89256111d0c88d49c93b34eb2791568120b1c5d","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"40f9eef31dd0f2deeae7139ade500078d79fdb61354a99e01e4f7d45a46280bd","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dbee3b1c109b4476264a6965e0c375a334b3d1fdb78b03bd3790eacdf2469ecb","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2487107","recipe_identity":"1d5f481ea767210f95a722307e8bda6ee98d4aa4f00a68dfebfeff9b0e02f8d5","loss":3.347574796598013,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"6f9de12d6c13f378eea100b8f8e98175d6851e3204b80c56b5637f385d27a7a8","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"95b3691951d0b7d7f820590868e5bb92d1f4e298bc0bae1f2f5b1bc7c1c96358","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"08394c6261bdee7e49c230520295bcdc12400cb6a40a4b642425a8543a3bae5b","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"68d32096df115a12d379806620c6a0075e0b5ca00c29b224dab9d40134e74caa","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"5bcd3fcdd1506a269bbb52cedae8cb1cffb569ecc40fe5a79fbc575be5ce94d6","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"4d7e71447b261625bd6768d8f777b633a92c434c5b92b3199bcc73a9e520104c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"cf323f04febf13a1b136795e5b62b378d1242253a1ced8a1ca65adde59ef41d2","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"910e80800db5fdceb87d3e3ba67b8a2fe01586701f712e8591d4bbfecab1804f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"efa816322e6d745271a3c697e0f2c7aca5e510c97be53cb1c2339db1eb15649b","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"a93897b2735369db29a0db00d352bb3cddd8f2a47ad39dfc9da0890b15ac1d63","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"5e120764f08797f2ac1f0a5cbe47a76e6b0e196622f1b6f714a9189a8dadda33","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"1e5410e971e8bdad91763f32cbbac7ddf57eb2201215e26b9526a4b9035922fb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"34e6a29a92ee2c5b83f17a8862cba3731df0d0b86455bb1d5a79b7f89eca6b12","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"b8528527037fc0ae3cbd7265d585da05cb0333c2798a73cab56fbc26f9c08121","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"c2c2d7210c244414dd43f78a0dd54d37505467f46d9b8902ded49eed7eff7bb0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"6bfbe5ec245c7c297cd06d2d17c15ab9ab3773d21a0f0df95e1d911f7a4dc8f8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"0071fe9fa269d7e01b5176ed098938cbc0d43939cfa33c0a30f6d6fe252d94e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"df39f29228842ebf91f0997327ff72982156e0a1db05ee5b20dafe7334b56eb1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"6808d7959bd5819b08cc963d613c6a1316b54afdb388ac3d33fe49aeb8aa4946","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"bb1255b75d7870bf59e94e9dc50cfbc35c0d5b151f168d904fedd0b2a7eed723","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"e6132c9c79ab7b1a9a6118e06b247345d9cd79810d16e74887e0c5e52f582b1d","bytes":5344}},"requestSha256":"22d81e6bfdc64a5530d740eda76936919d9ee13d8f8254da0b4316b66f1f87e1","metricsSha256":"86eb2cabcdbd2a78d2fe89ff5cd1366ef289227651742276b9d7452f64117169","continuationSha256":"4f843ebd56e07d735736f916bd99abd19b6a88b441a7269dd64ac542220aaecb","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"1e5410e971e8bdad91763f32cbbac7ddf57eb2201215e26b9526a4b9035922fb","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"34e6a29a92ee2c5b83f17a8862cba3731df0d0b86455bb1d5a79b7f89eca6b12","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"b8528527037fc0ae3cbd7265d585da05cb0333c2798a73cab56fbc26f9c08121","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"c2c2d7210c244414dd43f78a0dd54d37505467f46d9b8902ded49eed7eff7bb0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"6bfbe5ec245c7c297cd06d2d17c15ab9ab3773d21a0f0df95e1d911f7a4dc8f8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"0071fe9fa269d7e01b5176ed098938cbc0d43939cfa33c0a30f6d6fe252d94e4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"df39f29228842ebf91f0997327ff72982156e0a1db05ee5b20dafe7334b56eb1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"6808d7959bd5819b08cc963d613c6a1316b54afdb388ac3d33fe49aeb8aa4946","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"bb1255b75d7870bf59e94e9dc50cfbc35c0d5b151f168d904fedd0b2a7eed723","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"34d842ebe7a213511c5994edd97b7cdeb2c7071ad09eaa873c5833dc82b3e91a","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0c3a60b25c9262f6a013dae39ff66ab9413b6fd0137f10057971688b95e7d2d5","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1d5f481ea767210f95a722307e8bda6ee98d4aa4f00a68dfebfeff9b0e02f8d5","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2487108","recipe_identity":"9a20fca10acbbf5365c34d995961cdb3a88824c01fd9928c868022905b6a0f69","loss":3.358529524385071,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"943ecf0b0c269c937f7aa68f32eacc834d32deef1e612f116de813e575c74066","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"505655d6ad20954785735a85c9c5daef1e0a3f43a26c736f96072fedcf802e21","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"402d81150b9f3e29f8bff51be1d1ee66972a3b31a9c0fa47b30eff8d4dbd9f1a","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"cd5206fea629c8f034dbfefe4ddda777b5ee56eadfe80abb85b38bb8a410c25a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"b4aaa316531a7010bb6682596cf0fd80ee45b909ed89540fde208af901a78298","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"ae85ad2ca3db3202e7bf76223ba78b44b58a9199efa51168663f9b808ba04ace","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"edb88e360f04d8bcbbb3330875934e44f9d3273f79994ffad902724f0d427aff","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"d6f5c9600b252e6ea4e2ec30ac4f1dcc074ed4316ab232dcae28e41505b61b8a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"a1b51d5dd96728adc7d4bac26882f474aef345ae5edbe4a815384d1281c3f7c4","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"66acf039b0ffc0c403d38468e81d68fd6a77120f25fdaea7558f9a0da30cd925","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"c130c7c5903e4b31e296f1f554b8bf26966208a1f618778af033d40cd28c4239","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"8ab77603aebb6e35f3d63562687de24f8ecda661e6517ea2841c1ce524d3b9e1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"fa7d901f8b62d73d321d0d8a3ccd6ec09cc16df8c441af4eb61ebe28663928fb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"9d65f86e7c05548a235e38e3a986b2b12aeb6bcb5064706ec4bfcf84230ea229","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"f1c0b01695ac240f642cd3649c02f5e0a53b91560d1015bc71ab9e861e629364","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"a4f5aab1f7ebb40dd6def510d64aaf189e3bcd04f4708d839f78c5f2a6046b57","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"74c2cfb733a86877b2dbc2a03ce78900b13e0b44fb30c35e51ba7d6284085b29","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"f468f4dc728469d582321dcf7cd5a83debfc3b655ed911971cf5fd72a7dae4c5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"111e1a66bb476e9a63ad7ada579302ba1a9314b84769581900c9145952f0b409","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"2b84f104dd799c391f7524b9f7d079aefec91ff711882fc3154400d2b17fc9cd","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"f4b839ae078c794dbef02f23ac2067f1eb70b953b35c16306b79201f243eca4f","bytes":5344}},"requestSha256":"60832c94ae9cb1ece410a0f7514bd2e2a3788ae2e5ae4d195b0df9d403de2b3b","metricsSha256":"05d5fd55cf55ac349b0592aa532579d59a7c00094739d76f5ba4932d5bb4a9c7","continuationSha256":"8c351176e8f61d609c5887032212bdcdc8cb9380c406d53cd09b9ee33c653f73","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"8ab77603aebb6e35f3d63562687de24f8ecda661e6517ea2841c1ce524d3b9e1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"fa7d901f8b62d73d321d0d8a3ccd6ec09cc16df8c441af4eb61ebe28663928fb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"9d65f86e7c05548a235e38e3a986b2b12aeb6bcb5064706ec4bfcf84230ea229","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"f1c0b01695ac240f642cd3649c02f5e0a53b91560d1015bc71ab9e861e629364","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"a4f5aab1f7ebb40dd6def510d64aaf189e3bcd04f4708d839f78c5f2a6046b57","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"74c2cfb733a86877b2dbc2a03ce78900b13e0b44fb30c35e51ba7d6284085b29","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"f468f4dc728469d582321dcf7cd5a83debfc3b655ed911971cf5fd72a7dae4c5","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"111e1a66bb476e9a63ad7ada579302ba1a9314b84769581900c9145952f0b409","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"2b84f104dd799c391f7524b9f7d079aefec91ff711882fc3154400d2b17fc9cd","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"d0008c86a6ab0c01469738456b5aa4b27c69162a9cf7ff4ab228ce36daac3bc6","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e4dc93c704e9170517a38b072bd7c723a99e8665b42c1e021f8b133fefe33ef9","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9a20fca10acbbf5365c34d995961cdb3a88824c01fd9928c868022905b6a0f69","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2487109","recipe_identity":"c11e40c118f6a25831a29763617a834d8ac8c57c2a476518e3c8b653634a726f","loss":3.3577636468280074,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"d827708c87dc77b73ce292e774f96e8a975af1153ab4811256f4c2844d18f11b","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"e7be991a76c311473b94ea70a7527341d68ad3a7706f080f862f59a0aef7ba46","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"52c488e58c2031bf6169c3dba93da1079a3572bc1d1f6a487711d3f1eef51e20","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"6480166886f9ed66ca6a3afa19a2e3f229685afcbf92a35a9da43d3a2fb59758","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"f5ccd0af312fac59d1f4489057359678959844d93d510d42d34f3a250831acc3","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"f289068f46712c655cfd46bb443d055d83c616b14170e2321d1339c418839eff","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"fa1fe445c69b74b6ed54296b2e146d68d142349351b52b32b83e6cd20cf0bf08","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"79c21ca7b02de50106d77844e071724b61e62dfbd0809ac2940387f73b2bff79","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"6436e6df92c876ffdf16d310ec0dac5c525869eef96ea405f189868e85926367","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"03dd53df4f412ded817376964de9308a80cfb24fccd46394c86be62eb084b863","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"fc2106127374b3ce92fd87854cf8d6b3f56d6d50148a6c67d722c1dfa829946a","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"b80452cc4c17b9f27a15966c35f126bab964743ee7c8736e2eaea2a2dc73e275","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"62d3ae8e3bf1f82a9f39b0cb27427962d4c176b87c4018a26e0a64457a905e40","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"0a4b4d0477c3cf116b36b4a4edb11efdd5d8c4fcbe2196b29c1bb82e4ddc7041","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"156deca5ddbfc73cbbea7bbe5c68d3f4aba377bb39040e5e8e6737420f82ae0f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"4f12936bb4cb4fb8a78cae25e6d0101b309956a080017f003d6192c7e2fe84ae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"baf45ac573c1222af790b7684a7cfc8e917fd02db6f4a740af5f58e30f71dc7f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"a586abbf31fe8464a7eac54e4fff52caf10cc736f3bd023c03db962c433abb89","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"11fe18849797ddb069cb438d861dce92211b1babc3ab3f751948e88bfb27130e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"4d0f17500ac68db6bccfca057040a00657ccc07e02f11d0e6f894af2b61687f5","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"ee84e82f14e61870021489043f48ba49805c2207c3e3868cf1c8f709d4157667","bytes":5253}},"requestSha256":"eeb8738f6e10be258508165cace03dc6612ccd261823c57bc412887f17bf9898","metricsSha256":"ffe1b3620e401d60e1db45e9a63c3808b9606dce5ec7e357fa94e85a22e16e2f","continuationSha256":"11b145933e6ebb287a3be06314c791f869a71dd8b3e95a37b7aad2993e303fd0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"b80452cc4c17b9f27a15966c35f126bab964743ee7c8736e2eaea2a2dc73e275","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"62d3ae8e3bf1f82a9f39b0cb27427962d4c176b87c4018a26e0a64457a905e40","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"0a4b4d0477c3cf116b36b4a4edb11efdd5d8c4fcbe2196b29c1bb82e4ddc7041","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"156deca5ddbfc73cbbea7bbe5c68d3f4aba377bb39040e5e8e6737420f82ae0f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"4f12936bb4cb4fb8a78cae25e6d0101b309956a080017f003d6192c7e2fe84ae","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"baf45ac573c1222af790b7684a7cfc8e917fd02db6f4a740af5f58e30f71dc7f","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"a586abbf31fe8464a7eac54e4fff52caf10cc736f3bd023c03db962c433abb89","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"11fe18849797ddb069cb438d861dce92211b1babc3ab3f751948e88bfb27130e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"4d0f17500ac68db6bccfca057040a00657ccc07e02f11d0e6f894af2b61687f5","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"1b0d978c405ed9209a665e8915221791d294a7b42f39e4601bb8f82cce870f3e","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"afb0cd3807eb251d04fb89becc78a6b8ba8d2d8126f38d101f47f7d069bfadfe","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c11e40c118f6a25831a29763617a834d8ac8c57c2a476518e3c8b653634a726f","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.9--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.9-b2-0.999-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2487110","recipe_identity":"24c886aaed0102614e4a011752720f239dd5820500540122774b324f971621da","loss":3.358507563420065,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"b2ea80d7c1b434b03000d75ad7d47064c735691e35fb21c166153438e7d5305a","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"c3f4e9da5324f1fdf3a8c1d6eec892b697668a4837129fe0aa0b409ee6fc2ecc","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"939fba7a41e45e96dfcbce22f6c6921b9716237089ef50ffe56434376ffeab55","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"899f691877e073159a2f8e6114b17992351ccadec2535a764c0eaba3822d4578","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"12e8ebabf84029579688e346a0833eae776a60afdeb05d90fa51bdddcf199083","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"b22c854b6a44d0a4e6cab4e9eb9bfb628a93ecb625af593cc2043740c3485b5d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"aae25244bd54adcaf120989281738323023e6f54e04269c56fb607ae17cc48c8","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"c357286675d3e85e31e34c97c5c4aac8302e02d559dfcf0d63191fe90c219e68","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"d22df1c36f4992cf7631215896ab8017dc24536606af5fbd3038e73bd7dc58e0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"82d556f89f67b2519486654c55259d0481c42acf4fd56bb42594b199e586d730","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"00bfddd714f582c5819422c780853fba6c558b866f18c537fda01c22c143a361","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"d38cbef96511d8c6f0138d4b854ba393584bca24d34dc410fdf7fff8345a5afa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"e7411307926145e397e2b976b9c1d2c87339aebb0e1bfedda5af9059843c20cb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"fcee2a22cf9a0cde75520ba3b5c4e6ac70ef16202194763564545b3933c1755f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"2aba7a8d529e574c5cfbe8dae8539400cddc1afb8d4ab742ac7f2ab77402954d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"49aad23615aac1524f8ec687bdd3ef9b75f5186f82f380566bb3d844d67d21a1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"0d046dd8954c923816638bfdf0a189d2e1dfd3ced264bcedc83a95b835ebde5b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"565a1e1cba32aff2734d93ab527861722bc3ca37dde802e45c124aae4e516625","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"4474b100d90b904b845bda99f1b83b22380a791f79c0ac0e40b37d9165e71b90","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"481c91f9553a45dc4406ac187b3f89c3097e8475397a40469055b0def1302a17","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"1eca036d6b5163dc1e5a77093681213b1e4471c0641c98edcbb7419a3ebe52d7","bytes":5315}},"requestSha256":"36f19959262d39a89c37f738cafc382439c926d5c9a02f30891db7dd14deec7d","metricsSha256":"fa01c6c517880161afc450337d9e2f4633d44b1cb049d0c349cd6e401d628a90","continuationSha256":"0909a9bb11898d1d7cbb3daf6d8291d629632eac9ab0b0a22c713e9dd2423032","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"d38cbef96511d8c6f0138d4b854ba393584bca24d34dc410fdf7fff8345a5afa","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"e7411307926145e397e2b976b9c1d2c87339aebb0e1bfedda5af9059843c20cb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"fcee2a22cf9a0cde75520ba3b5c4e6ac70ef16202194763564545b3933c1755f","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"2aba7a8d529e574c5cfbe8dae8539400cddc1afb8d4ab742ac7f2ab77402954d","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"49aad23615aac1524f8ec687bdd3ef9b75f5186f82f380566bb3d844d67d21a1","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"0d046dd8954c923816638bfdf0a189d2e1dfd3ced264bcedc83a95b835ebde5b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"565a1e1cba32aff2734d93ab527861722bc3ca37dde802e45c124aae4e516625","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"4474b100d90b904b845bda99f1b83b22380a791f79c0ac0e40b37d9165e71b90","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"481c91f9553a45dc4406ac187b3f89c3097e8475397a40469055b0def1302a17","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"58f89524d33a39525d3c3c6f2c2ceecd5834856f119ca04920a54446cc41359a","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"2bcb0ec3776202a9a5f3de54d3107f13cc0f0bf929daf7a882cd6c97f35c1b65","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.9,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.9-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"24c886aaed0102614e4a011752720f239dd5820500540122774b324f971621da","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2487111","recipe_identity":"944fa97d6eea38485952d5cd40148cfefb3485a99741a705791da1df91687425","loss":3.437422778839181,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/result.json","sha256":"cd418a192ab18ffe06da684bb8b94918983c3fa55781af93374ec67e5dd736ec","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.pt","sha256":"7c3c4567119710724e1250ec98943414a11a90d66e5d6db806f349845ffbfdab","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/final.safetensors","sha256":"56b65ced496329c4cecf49669c5a0afed03dda034253970aa7c81723b11eb220","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"74e98a34002e266c907644d5833536ec8f72b9e740f0686784575c2495a3b3ed","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"779eda31ba4e300ac362435e90632da67d5b04d1ca57d9f0aeb0ef489345c985","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"164f6fb89b84f77a28a32d6371a2fe9c0e49dd321cd103e5d0871b5388a4514a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"10a8f04bba06016b6b4bf65f6d0abf0cfa015a4a014e13eba4589572376255dc","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"266d4b7c173199d6d2d60dd263679d90a32950aff6c8f670f5128a30c5913a29","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"d6cf76a0b8f37d1062b5c374ec8f48677cfd0152d759e3411a09f81e754c9945","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"e7e7b33e1d6f854583d6bae58664461b111a174173c1b528b38ee48ea65837f2","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"b8cab428e4c0c5054839c15e88c08a630297a091c01f1e9ef91b378089419c4f","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"64e1237c5773790ac198606b544035c565733565b518f5cf24c9cae6f4e50bf5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"09c0aced6da6bdcb44ce250a3993339fa70a8b5331a252970924a53df65d0534","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"e4bd8f5a82bc647ba954536387a71fca1957cbebf52d075082721ef78dee59e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"5ffe3983e91758b7f2a0a08da92ba79589c83eb4a6ddf65c63cc456d828616e8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"ca3b9716d3023776ff41997e5cf0b6f3d333f39343841c558ede83be739f30c4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"2b2151820f76164182dde5e90fcdaf91aa9ddff10bf183465b3008fc57c194f6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"4fe18744afd7570c5a016a9e088af0ca2f8e75f881f349c1416afbdf2f26395f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"0a973f2e93479f28946d2ac77657b8d189a46dc9263da75c3b39b4f260754322","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"686f6040abee4bd33b5e038be5c39d71ec2e139016e55f7992e81825c1c5377d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"4063977e469343ae080ad4fcf98f18b3ffc3a3dd29306e5127d2589c232d82da","bytes":5344}},"requestSha256":"51b65397e52cbca5654159d102f8b823ae3676a4e5d12c743dbf8e7542f827b7","metricsSha256":"c0ba6f9cd78810cbc82e2caa0a226f533292a1796730fa4e988396079fd1221f","continuationSha256":"7fc632793c9592a422575e9f91c839e9f5aec60f16f5f730b382c9fb0986eaa6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"64e1237c5773790ac198606b544035c565733565b518f5cf24c9cae6f4e50bf5","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"09c0aced6da6bdcb44ce250a3993339fa70a8b5331a252970924a53df65d0534","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"e4bd8f5a82bc647ba954536387a71fca1957cbebf52d075082721ef78dee59e9","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"5ffe3983e91758b7f2a0a08da92ba79589c83eb4a6ddf65c63cc456d828616e8","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"ca3b9716d3023776ff41997e5cf0b6f3d333f39343841c558ede83be739f30c4","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"2b2151820f76164182dde5e90fcdaf91aa9ddff10bf183465b3008fc57c194f6","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"4fe18744afd7570c5a016a9e088af0ca2f8e75f881f349c1416afbdf2f26395f","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"0a973f2e93479f28946d2ac77657b8d189a46dc9263da75c3b39b4f260754322","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"686f6040abee4bd33b5e038be5c39d71ec2e139016e55f7992e81825c1c5377d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001/continuation.pt","sha256":"a66b347b04abc26ce0bd3d92b3c47c229eee329928cc8383d2454d5ed4ee8198","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d3fb75594397f749f86faaac9d8c6a9441c757aaa1299c24c52f625133a37ef9","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"944fa97d6eea38485952d5cd40148cfefb3485a99741a705791da1df91687425","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2487112","recipe_identity":"1ee5623d844ba1636050b8f600d355b996c76371ec589481aafe7fa309664ac4","loss":3.43905257227758,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/result.json","sha256":"5170ba50d76a75e4e4664ab9c4d92aa5bbfb20331e80888ec00c2bff330bf935","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.pt","sha256":"988f6321013c8f29878d6951c5b58f6cc3c15eceb6e7fe181898a245af146279","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/final.safetensors","sha256":"a12039a216473890593b5a6a55d5fb49230a49d1729159a0310d710591b7b8fe","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"75475e2c891d94c9b9947402263a4071d617800b0a4565cdf957581a28dbfc94","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"ff271b299e306d9ae4dbe6c1df7b37d5c8c70b8ad1b74e93566d6ee0d9ccb427","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"026b5314e605c1ac74f2195b129378095b3a11f9ec60ffcffd549802a35c5f0c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"e5fe11c3cb0d14a71f53a805b4ccd2b3cb506688ad5886e631ce263e4c38a610","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"c1682758dde1eb2e7008ead297f560896921a2332697cc9a0ceeca2899eb8f5c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"06d4a51532bc6204b19838ac4bfe0a794ed46aef5ec6afa2be3c7af6e2734190","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"ddca0d738195837870c529744bdc4e28942c2831783fad8cc68919359732106c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"3176ff74eeef278ce20ed81cc11c5e4ed36486e39d3e5827ae933bd018cb41a4","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"ea4b7a5231e24146c694622d8d05800d43e9572fdbcce0f7b77667e0aceb7707","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"2d9004f0ee8203f6e82fe4cb957714d5a2613c866ddd8ab95c0ba3a487e05bc7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"b08c68e8854001d82009487af7fcc222c2bec1d7d1be586751ae613f313e8b5b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"4fc078cb87d646181b86f297d336cf30c38d0f9c3a166773090dc242d83b93e1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"fc80dabb1782e568a9fa86d802d00262d08e2fe75bce9fb40a1ea40f0c716ace","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"8d14de57e40906f0a18d80b3373db847f7a1ceb2009f49eb64ed6217b6b7a635","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"624777078a7a3b2b5a0c883984c71dc6e544d073fa5cf5f5ed76c4692a1ed3a0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"88e27dd1769b117b8a4dfd645405c6781dfa0e20e6193adce741eba07f1471fd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"a2a2ec8bd1400a99965482aad22023a99adad8e8f75b207c114eb6d99e555500","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"a1646891b4be1da1e0308a59e09963d8cdc78b8b1616bdc79071a8965778029b","bytes":5344}},"requestSha256":"5010245272f23b881e2806ce86d11cfa9c7577b3123e426265fd1fb04fb5f4ec","metricsSha256":"9283d1eafc2296d22f0b2795cdb87e59fe5716115e20378ee32a938fba3ff90b","continuationSha256":"dd37fdd410c0c732cf08f855d46d14bc0e786c43f45c93e87e686864c6dfd785","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"ea4b7a5231e24146c694622d8d05800d43e9572fdbcce0f7b77667e0aceb7707","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"2d9004f0ee8203f6e82fe4cb957714d5a2613c866ddd8ab95c0ba3a487e05bc7","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"b08c68e8854001d82009487af7fcc222c2bec1d7d1be586751ae613f313e8b5b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"4fc078cb87d646181b86f297d336cf30c38d0f9c3a166773090dc242d83b93e1","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"fc80dabb1782e568a9fa86d802d00262d08e2fe75bce9fb40a1ea40f0c716ace","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"8d14de57e40906f0a18d80b3373db847f7a1ceb2009f49eb64ed6217b6b7a635","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"624777078a7a3b2b5a0c883984c71dc6e544d073fa5cf5f5ed76c4692a1ed3a0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"88e27dd1769b117b8a4dfd645405c6781dfa0e20e6193adce741eba07f1471fd","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"a2a2ec8bd1400a99965482aad22023a99adad8e8f75b207c114eb6d99e555500","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001/continuation.pt","sha256":"cab1f8f4a5b12d9f1413941ee81b767426959e9638fe43e1b5ba1017eaedab01","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"dedf1d797344e95f44f5cd1d265acb62189451c7fd4a6f1085cb88f67a8c5bd5","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"1ee5623d844ba1636050b8f600d355b996c76371ec589481aafe7fa309664ac4","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2487113","recipe_identity":"8b61a66814f097697b33e9347dbb06c919ed750764d35cdc89cfea64ad81968c","loss":3.4385674619181477,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/result.json","sha256":"8d9bc1409cc48e12dd9dda9eb52e587bbb5b62eaac3e5d260e57ee1e33b86bd4","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.pt","sha256":"104ebf52c4af47fb1c40954b070cc31f4f550d0291ebaf2f40ba960243759e8f","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/final.safetensors","sha256":"b98526965a7d358b30752d2e9df449ca9e5cb8b2cdf7d594f0ebaa0fdbde2d05","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"bddba2c29f01fcecd07ac6cac6a39195efc1e0cce119424aca9cea952240a706","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"4abfe27a21856b81a6cb2ee24e4911cbfcd26c74a596f5204c5c3b29fded6373","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"ec5ae3616b6225045b0ed091626cbb3d165a93c119577be0b7779c29cfc64bbd","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"1e75a5334b4d989a292690ccb3a3cf0265fb1b919b330eb25c1a222b8086dc5a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"a6c2ee660336602f40091e76d8c894cfd42dab012ea5b7d9c3ead01deaffb160","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"595291b101154b9b7e872e21b90c32771bb5d1830e9c6211865c0400bea63838","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"dc37c6adb24dc7be739f9155da9b303b8ef5420ac333850baeec4c3356cb6984","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"1a56292213bafc2b48bd7de1221782aa54b8ba17304dbb8539c421310785ec41","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"3c1241fac308d4d5ebf5a3e070d808f55909437dcafc36119bce5787f6adf22b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"f79b1a4db7fd5e4c98b23c554e6e12cf500e3c4b5b32a9e613624bc8a2e21bd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"234937e4e816599d1ef9d8be977547dbae4d9d28cc27b04cc616f5d51da6b8f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"7039d85e3433834341cf52c0627d405b879e9c15a8c65b10f3af4ebe53d480d6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"315ac57ff80d87927e1bb591770a3c1118fcbafe83b6cbbec4e98055fd948e2b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"e7eb686f47106193bb53103b316a22af0b024ec53bc31e2ebbe86deb2afc28d4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"74b29d941f72845865413751324b8c1b5233620af246f22a350d55797d5ce8d1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"aa93d77dc8693ce9db38b754d0b21bcd8585fa2b89a430d1d1ba51cd19075a97","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"e7a6cc8af5e889083c59088c1f3ccdc2330960d0c7e02c9e631d564a1f9f3e30","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"be9b0979d472c923091921678668148a1feac67026827a58be29cd7334e5a6b4","bytes":5343}},"requestSha256":"29e6366dfa21976a3ff35fc8beeef4f5df83f6fa6a3993aad173823e9caa8ddc","metricsSha256":"d2c6d0c93d4573b64a805729ea50afd1021f6003b766b9d40f6b057da4bb59c2","continuationSha256":"eaa1b1014d4e0eab02d77712f4641c6e469a6d806f661ccdd32cc8983c0b993a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"3c1241fac308d4d5ebf5a3e070d808f55909437dcafc36119bce5787f6adf22b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"f79b1a4db7fd5e4c98b23c554e6e12cf500e3c4b5b32a9e613624bc8a2e21bd3","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"234937e4e816599d1ef9d8be977547dbae4d9d28cc27b04cc616f5d51da6b8f8","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"7039d85e3433834341cf52c0627d405b879e9c15a8c65b10f3af4ebe53d480d6","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"315ac57ff80d87927e1bb591770a3c1118fcbafe83b6cbbec4e98055fd948e2b","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"e7eb686f47106193bb53103b316a22af0b024ec53bc31e2ebbe86deb2afc28d4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"74b29d941f72845865413751324b8c1b5233620af246f22a350d55797d5ce8d1","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"aa93d77dc8693ce9db38b754d0b21bcd8585fa2b89a430d1d1ba51cd19075a97","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"e7a6cc8af5e889083c59088c1f3ccdc2330960d0c7e02c9e631d564a1f9f3e30","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001/continuation.pt","sha256":"c8a843175cf49dcb32abc4c76a62db475179a1800829432b84326ab05ce17251","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4309c1d25bb5875d3c3fb05cf12be9850b6400b3bbcd07ad62d6916d04312699","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8b61a66814f097697b33e9347dbb06c919ed750764d35cdc89cfea64ad81968c","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.9--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2487114","recipe_identity":"855f7c6f85eb08b7f5e23751202eab85c114bbdf0227d1f9ee0c0f7bbabd3b2d","loss":3.441580831168483,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/result.json","sha256":"0d1d7cf9be4c5f0f5e6f62e56629d143cf972e2ef57bfed67983b01cce6f03aa","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.pt","sha256":"0af9ba1d615dadf0fe34735f384bcd22edf624ad9aa4902e91ee9cdcebd3cbdd","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/final.safetensors","sha256":"47b95b41abcf59a8a2264754211204a1c4ec8f3fba3e890355e9a815be5acae5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"60b808ca6d82248805a041a21a8704f7c4a2f89a8e6db43b3c6f1c7fba6f7f9d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"279afb0092fc5dac9d46f54efb840b081fa84202642a945ab1f6e59f095ab8a5","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"e205e297ced873047382b75a0a75329c69ead63b9966c7156635ed7b9f2e2cd2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"b48eba324c75106380cc5685adc4d722b040851593e920004af9373d5b2efc39","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"0259d7685dceb36832e72736b79c9a4684ba5ca76a843a3dd0cdc2181b3530e2","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"23f6df0469c1fc144d701b05a982601627919f06c3437a1586a5d4bf6e0e89b8","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"94e763be590f69d6f709621d67e7f7b01b54c3887805058da74241176668c63d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"9b5ad774aa1ff34a6e5d182d6c39bd3d6d254b59f8ef9df764642ffdc79410d1","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"f3bd4780c16339fa56bf6163c7db76e8a1847c2af220c2c7c3148037dd5c29ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"2beeb10b84bb1f240de4627fc5ecbc795281145e1bac528f81660f47cfb4763a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"8e8207e01c54eb158eab1baea6d593c94b634912a8e0941f775fb7702e573719","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"2cbf6724258c3aa889659f14b3101a0e7497fb8a20e17811e97b409acd6d16b9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"29af7b47c136503ada4797b288c4dc07b6925032e73d276423c2f8d103adc849","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"c363ac2e8a9c2015fcbfa6a38b6811f425d8b6a47a774b79b335d9f55aea98f4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"9ac07d27a5643c0bbfcae1ea02b0dab0545b44a2f29760710b3827098046189d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"fa3f985d321c144dd22d9673ac957a5bc8c2f432049795a1d6528532f8580dfa","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"3884b148379b3d7450d9b0189c56d933942919f9f3c4db0e239e98c20f26d0fb","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"17d7993dd02f4cf2a057ba7570c75e1960d362cc41f19c970f51e2051da363e4","bytes":5315}},"requestSha256":"f6b1ff9fe3f71a42214a889dbeef277ad77be121354c6763e3b064d8e6a233b5","metricsSha256":"1363109eab239f376943c581e2d2b3d9f01c57b8062349ae20ca41478c30b9b2","continuationSha256":"3330b72d6d341329747da63b5a94ad8333480618fff3e37d57f15c6295c8767c","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"f3bd4780c16339fa56bf6163c7db76e8a1847c2af220c2c7c3148037dd5c29ff","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"2beeb10b84bb1f240de4627fc5ecbc795281145e1bac528f81660f47cfb4763a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"8e8207e01c54eb158eab1baea6d593c94b634912a8e0941f775fb7702e573719","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"2cbf6724258c3aa889659f14b3101a0e7497fb8a20e17811e97b409acd6d16b9","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"29af7b47c136503ada4797b288c4dc07b6925032e73d276423c2f8d103adc849","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"c363ac2e8a9c2015fcbfa6a38b6811f425d8b6a47a774b79b335d9f55aea98f4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"9ac07d27a5643c0bbfcae1ea02b0dab0545b44a2f29760710b3827098046189d","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"fa3f985d321c144dd22d9673ac957a5bc8c2f432049795a1d6528532f8580dfa","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"3884b148379b3d7450d9b0189c56d933942919f9f3c4db0e239e98c20f26d0fb","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001/continuation.pt","sha256":"9ec3f93214db6e32c3e60c668a9b57ea66407d0a80b4a54116fabf03bc58fa80","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"16144d3a7f152c2daaa9b461d29985e1805916d5b3fb1ac8dfcf0c5605c6c439","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"855f7c6f85eb08b7f5e23751202eab85c114bbdf0227d1f9ee0c0f7bbabd3b2d","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.9--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2487115","recipe_identity":"75427cd09730007e4bbfdcd62d06451bd11e607039e5f96ff1f3b6dd0cc1295c","loss":3.449206524816929,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/result.json","sha256":"688f6103e33f350961f54f3de050b7d1a08d2aba822ae5267cc4642f5ed7d094","bytes":888},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.pt","sha256":"fa1837ff3552e1759183bb1e8be3f225ca8551ff9f4a468ab988e246e628cfb2","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/final.safetensors","sha256":"f5b10217a0f617a361d7b4526166a69df12d2988f1a489de4949356bb99a63a3","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"c201cffbd4d7eb8e1544225d102b5934a10746de17b654a7f50b6422528b5814","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"26e911ea325a0b71773f961a203be78521a5d1950e7a59340dbeb30555233473","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"032f1cbc09250b82e91676a08fffa081c36bde9c1d2bbff16f52cd61634908a8","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"e1255c20b4d449914246d5f91a4f313b49ffb80ee80210d6065735e555f973a5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"d67cdc0e00fbc21cfeb2e052e746b2809734c07dc31e14198109804d4b8eeb1d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"9fe30df324298a68ee06a3aa59511907f7038fbc4d4df2607bbdf38931b76ec0","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"7a1c3fdfa8fcb85c8605baca0357b9c412c582baabcafa05e796fed12b06d90e","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"8485814a8abf0236846c968352a45de786178b8dd13a72a7f785538a89473a36","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"0a5aa79c736787b79c17a896110908439c0099c4fb29d401aeaf400748b95316","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"9d4d78014bef6ef473a2c541eb48b162365b77c42a43975cf288e6e436fb9472","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"3001eb4761e108d09644d91190f47ef89af33c0adcab9c32adc60a5d4357696b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"e7b9bf7315a6fb264f37a7aa9367b8c3b962ac37c4d01cb93048911a8f7a1cec","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"d25b0ffb0a862c3c2a23636932cb7717868c17979a8cf862ae92c2d4867b1ff8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"fbfd3ae7ede9bc903ac693ee00f35a1fc3dea9a82baf4729709a700f9765e6f0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"95800418772b4f1463cc1f624d86554e395ab0d8369c0f18291b7e9e385f28ed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"1dde115b1706bed9e6ca68c258870e22ce7359346bd33980fa78ad9365968a4b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"f8eeddad9d0165a967f2609f261578ccd5ada964767470890d3bcd5a26c2b666","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"ce46ae4b1f9f92ea7ef13e6acca37c8d89fe656d88e333e3f6b58cd625f81144","bytes":5341}},"requestSha256":"1feedab8da24b0d25800cee1c8dbb5457246c1b7304dfe79aabacf0357c2b927","metricsSha256":"c68058d7bb6d01a2cc9cadcf5cba6e42af9b582f90e05fc4b995ba6cd93e2ac8","continuationSha256":"c39e05bfe0a7378ee26f5ce53402f922e7857dae92bd952436eb3ce21cbf7d13","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"0a5aa79c736787b79c17a896110908439c0099c4fb29d401aeaf400748b95316","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"9d4d78014bef6ef473a2c541eb48b162365b77c42a43975cf288e6e436fb9472","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"3001eb4761e108d09644d91190f47ef89af33c0adcab9c32adc60a5d4357696b","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"e7b9bf7315a6fb264f37a7aa9367b8c3b962ac37c4d01cb93048911a8f7a1cec","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"d25b0ffb0a862c3c2a23636932cb7717868c17979a8cf862ae92c2d4867b1ff8","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"fbfd3ae7ede9bc903ac693ee00f35a1fc3dea9a82baf4729709a700f9765e6f0","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"95800418772b4f1463cc1f624d86554e395ab0d8369c0f18291b7e9e385f28ed","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"1dde115b1706bed9e6ca68c258870e22ce7359346bd33980fa78ad9365968a4b","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"f8eeddad9d0165a967f2609f261578ccd5ada964767470890d3bcd5a26c2b666","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001/continuation.pt","sha256":"ee88b21f8ee7af94cccd823a7f85f3e842cf412c0e9329aaf0a1a2dedb561b31","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8a260e64d501acfe57c71485ae753a6489e8d722eaa5eb9a519451987bb28767","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"75427cd09730007e4bbfdcd62d06451bd11e607039e5f96ff1f3b6dd0cc1295c","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.9--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.9-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2487116","recipe_identity":"79b36b6613ede1817bf65ab1d1a8fde69174e5ae4b5db056994b39ab727f75b6","loss":3.4497888872279514,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/result.json","sha256":"752263576f8eeddb5c45d6d80873675b357a5d4bf9f7a397ac13f7631cf1a722","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.pt","sha256":"3172ddd90238642e1946db77028021ae0fda60d8e6bd37f2aa64407e2cc0bd41","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/final.safetensors","sha256":"14bed17df649365b0d8382fd171afd99b2ce710ead5e33b12f560f6cc397fb17","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"93978897a38ab6903c6624f0dc50ea9bbb78ca75d16f11b5df46fa0542682dba","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"a16c2e804f9591b624c9913b999a556b66827b014b2218f70202156a9bf12ee0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"8c1ef03e82fd8363a537a463e0c9f60a275f55650cdb7952cde77072d7ac205b","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"6d3c242d8006fa6054d85695410b7c1bffd1ccb74e3197049f33128b5b1f37ed","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"1de4378cd1ea0738fca0fee33abf83c4384c6164aa2f31827e544846cd932e5f","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"034ff76b8993b50b0b9c9ce04bd33fa567f2a8d38ae6cade4d3f8af5f488a3cd","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"14a0926432d66848a1178533ac18f070d0a815dcc60bb6c86671e55c149b438a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"0e989ff50ab273b5ba2c68ef96e4c3c31fe81b0b68ab70cf02e46c6f9d2258fc","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"e3f8551385660c8b912d2a68bd28da4f8f658cbcdd25df66ccc29ef0c6ab2dd1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"9a1af2575d8a564e45a6e2f48fe283a4f0dd17d08ecf5043cf168f23198107fc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"d3030a27f0906cb254744bc3216c836c96bd7b8ce4b889777c3a3877d6b6f879","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"aa5ec237284cb298a3c07e4aec098d16f4e71593a1dc0daeecbfbedaaf524c80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"86b5603fe1135c26a21de2e337dff880a4bcdc4b6f6b686e3153aaeae945cb09","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"02e6bf341529294b5dcc3ffbe66f8523b78ae224959dfc7f9886267992be35e5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"867569e69559067eb979105749c1a93d136ff76924ec7ad5a80bcde6aca58b21","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"195affec030adf84bfacd483649c0606c8bb57ca8e157dd1c081a43cacbe0913","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"f9c5d1e9498be0a6903edfd5cdbb2cfaa016ea9eda703243212bb6caa08bcd76","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"9bba1b406c328e324c389e1e92159ba79b7061fc488709ca645419045aadffad","bytes":5341}},"requestSha256":"1146ada5f42b9dcbdbc92d017e3902eb38e36f7670998c73ec97474c412c82f6","metricsSha256":"d64cbc6cf03356a918644db4fa81834192151d3251be4882f998026305f0fcb6","continuationSha256":"037de2251bc25c23f2188394fbfb37e4c9f61148cf264f82e0d66ac3f5e3ee34","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"e3f8551385660c8b912d2a68bd28da4f8f658cbcdd25df66ccc29ef0c6ab2dd1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"9a1af2575d8a564e45a6e2f48fe283a4f0dd17d08ecf5043cf168f23198107fc","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"d3030a27f0906cb254744bc3216c836c96bd7b8ce4b889777c3a3877d6b6f879","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"aa5ec237284cb298a3c07e4aec098d16f4e71593a1dc0daeecbfbedaaf524c80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"86b5603fe1135c26a21de2e337dff880a4bcdc4b6f6b686e3153aaeae945cb09","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"02e6bf341529294b5dcc3ffbe66f8523b78ae224959dfc7f9886267992be35e5","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"867569e69559067eb979105749c1a93d136ff76924ec7ad5a80bcde6aca58b21","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"195affec030adf84bfacd483649c0606c8bb57ca8e157dd1c081a43cacbe0913","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"f9c5d1e9498be0a6903edfd5cdbb2cfaa016ea9eda703243212bb6caa08bcd76","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001/continuation.pt","sha256":"01d1a5edd6f9027f57a2c74498c6676f58f2d95c55850813288c2106ff31eb7a","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f220667c2cebcee4598446f10850c61c6309e2588f4e9976e2d6697446bdc424","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.9,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.9-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"79b36b6613ede1817bf65ab1d1a8fde69174e5ae4b5db056994b39ab727f75b6","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2487117","recipe_identity":"837bd634da48e8c8e67d833568f6f3643aaf9de7c4c7d0cf76ac2933717a67e7","loss":3.3660593553930678,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/result.json","sha256":"acc9ff0ce089fb1100ff522c3cda3ae5baffd4913d5eb299bf57046b89a39b00","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.pt","sha256":"eadc61134e2d8dc8f7769ef98cd172a0411d558882e3fa8cb0c76b0b961fd512","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/final.safetensors","sha256":"fa6d08e2e733e24ff33c3205a49b977a88c2c2c6e4fb8e28084df54fca562740","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"5ac6389b5445d03ac31faf8c08935ef0ee36a162bed841487c99e2cf77e573ad","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"cffd9364a8dcf54ef5c9dc51f8cc1e246f4e92d47fd525caff53c252d679354a","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"15dd15ce3ade8cc30c25ca3bbc06386a24101eaff403088277fd8a34bb7c1b62","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"168dc0593ebbfaf72fdbb976bf2de7b0c197306cf554512f128c9921248adfce","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"0dccb617cddf2244b1b251c1b39d972c499ddc95d3e689fc87773eb2d084380c","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"7b40804fe7d4f7be1127940d40d41ff842e6de32d9a39476eeb318b9f18154ad","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"4bdf627459709a3708450ef849102bb356651646673126cc541b44d2ec923cd5","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"d81391eada38f155967a5277bb9e4931946cbcc5bb14bce9ddc88372724c677a","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"ada3716eb7ac32ce14ca6411cda13329760503071ebabc8c6f49a7112fe171f1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"26ba7c0430ac496ceb043c286339be272cd1861771b611203fceee2df1dfbe4f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"44719a99f0aab38868cb77b75ed8f219a76cd8d953aa100eda736efb3a1f24f6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"47d4c734fca4ad88dfe44cdd1d278e2dd583572e15a4747ff54a4d2e6d894509","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"6a9ac3eda0319e8abd486bc1d5e8e22aeeb2dc19298be024998592a4300ef418","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"54e8d0f45c6fe009eb026ec20570c1422216f907c075c65e6301e30d8fb96c1e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"2daaa7a3095396f151125951266b094599c50a36f525da908bbadeab396cce85","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"5ab449c326196094328f2e7ea51048573b4f1fa583320619e23364133a7258c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"86d23a8c3f3af5af0d0dcd0a2bc4b3aac744e6e29e205bdddaea9620b6f26e4d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"6d4dc9b846926296e35f8174843f6a03e62895ed76e17547f06e4d7cee4f10a5","bytes":5344}},"requestSha256":"ea4d53cd9694924a9426fdb1d32b722530e634fee3287937c05ac89296c261f4","metricsSha256":"2580be2c1477101719ccd239b9ac947fed89b6083ddd536993c6202bbd01e0d0","continuationSha256":"b9ba19363f8c2334908d46fe509113296afd3ca01821aba3d42f47ca186493a6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"ada3716eb7ac32ce14ca6411cda13329760503071ebabc8c6f49a7112fe171f1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"26ba7c0430ac496ceb043c286339be272cd1861771b611203fceee2df1dfbe4f","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"44719a99f0aab38868cb77b75ed8f219a76cd8d953aa100eda736efb3a1f24f6","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"47d4c734fca4ad88dfe44cdd1d278e2dd583572e15a4747ff54a4d2e6d894509","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"6a9ac3eda0319e8abd486bc1d5e8e22aeeb2dc19298be024998592a4300ef418","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"54e8d0f45c6fe009eb026ec20570c1422216f907c075c65e6301e30d8fb96c1e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"2daaa7a3095396f151125951266b094599c50a36f525da908bbadeab396cce85","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"5ab449c326196094328f2e7ea51048573b4f1fa583320619e23364133a7258c6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"86d23a8c3f3af5af0d0dcd0a2bc4b3aac744e6e29e205bdddaea9620b6f26e4d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001/continuation.pt","sha256":"bdc2e7dd5ee9327ad0f39392ca647fcffc5322915504d21435de00eecb858ea6","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"c28fc938b8e56f9bd5d55fd60708f0da25825703e15fc4e8e7b6fe34c75406bf","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"837bd634da48e8c8e67d833568f6f3643aaf9de7c4c7d0cf76ac2933717a67e7","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2487118","recipe_identity":"a4c1ed5169056a99c5c196c41414612767e792485941404cc80c8bd6520d255c","loss":3.3746523820282865,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/result.json","sha256":"378055f045133edb552ebe818a2ec294ac5e0f28e40c48bf4d713bd5bedba486","bytes":891},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.pt","sha256":"4c2ceeca6d203fdc728e57f56dd71b7cfd93500d64f26701b249149eb0405cc6","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/final.safetensors","sha256":"eb4d95b1453d1f49796469b63aea2be1acde81fd279f40a4924c2ba04cec586e","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"3e2319cdc7c99de420c11996cc934cc8ce54cd2730c4897e87212d2015f8d32d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"00c1c512649bd843820f8f72e7b7a8fffbe74ba2509f425444a64bff39217a05","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"4f97fdf51ef9d4835624dbc49d02d39ab50f79d38f74e78f4c17085faf283247","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"42665eddd1e85adbee9ec9bea9ae3ae5617343aa1dbfafece2ceca74ab272b86","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"5548cf4c7d6f87d22c2cbdca07a89aaeafe95cb9dcc17a8f0fa3fbcf3d68d545","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"21b6c2d75bb5ce97a4f564e33b5f4fed60574d57de0bb6646053eb722e01939c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"4fe30b42729373ec1c8df529a8b9f57e0f6c80d89365b27c1d65f45db5454074","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"c25768a1190e753c0394031c441ba222b223c5de6e5be2f2b34feb92d8064525","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"7e83ae2e8011c08296eb234fcbd11c2a428ef74d28e5db9f356d16baa40d7e31","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"2d17d059ef49cf982389bebec7e59c0f8def0c9ec421d98086e6e110ee2c79b9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"5cf984f53a27247bf902a1c389ae298a1f310e15f05061826c0657ec7b8e40e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"b3e28d8dfebc42d59f4f780a0f199c9368d91e0b9df6d65c31c1c07e56e8eb92","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"514d8aee6f1aca58df22e14a1bd4f823ddbaa10086472efdc8e0ee333a429e01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"aa0f4480f46678150945baf18c499501c8ce5aadff55ea3ff6bec5a2143c7b3b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"2100db948ccb37c1e396d1d08a60ed82fd9caa026fdc921b2c138f4c17217e25","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"d0c940328dfafaad39c00943749ef5f650cffeb91ebfb2f4057f71cfd5db66e6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"7452b1c8355a38b12107ec1c674fb7ff3d160a948b29ef3c04bdde9209b016d6","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"ee92db46000534109df83db247fb479937301108dbc495407dd676c309dca4d8","bytes":5343}},"requestSha256":"f5c2ef37ccfc2d5360f2515af204268663dfb1ef366086cb5a968a87bdb7ee10","metricsSha256":"9886e70430d7563b5a4147847a526fa346183a1b632b475d734e42bcb8f09ba3","continuationSha256":"a069ff6b5fe04144761d3e431ee895c1a3719daee8261202e8969e10b4c7c820","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"7e83ae2e8011c08296eb234fcbd11c2a428ef74d28e5db9f356d16baa40d7e31","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"2d17d059ef49cf982389bebec7e59c0f8def0c9ec421d98086e6e110ee2c79b9","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"5cf984f53a27247bf902a1c389ae298a1f310e15f05061826c0657ec7b8e40e5","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"b3e28d8dfebc42d59f4f780a0f199c9368d91e0b9df6d65c31c1c07e56e8eb92","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"514d8aee6f1aca58df22e14a1bd4f823ddbaa10086472efdc8e0ee333a429e01","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"aa0f4480f46678150945baf18c499501c8ce5aadff55ea3ff6bec5a2143c7b3b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"2100db948ccb37c1e396d1d08a60ed82fd9caa026fdc921b2c138f4c17217e25","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"d0c940328dfafaad39c00943749ef5f650cffeb91ebfb2f4057f71cfd5db66e6","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"7452b1c8355a38b12107ec1c674fb7ff3d160a948b29ef3c04bdde9209b016d6","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001/continuation.pt","sha256":"31f636c90921b2d9505f1bfe98277ae11006048ea80e0054d0aaf8755be4721d","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bf5c6c37216ee4f346976b7c06ac2383f6759a52513e8d5736f38ac428e5f3da","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a4c1ed5169056a99c5c196c41414612767e792485941404cc80c8bd6520d255c","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2487119","recipe_identity":"08d0b1068a4f96e20ccd32695f9a3452634e75d2b7915b8de80c64b2f079d1e9","loss":3.372256653384429,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/result.json","sha256":"8156b44232f2821096f77ae4ba087477c32ada243b0beaa1eef0fd12e6879f81","bytes":887},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.pt","sha256":"5a12112af596e52fc0b7b0574d344edd4323b6b765c74a4a996f6d0a5ee3b777","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/final.safetensors","sha256":"0c0ccffd9d3efbca82a60c1008d6b2c03cfd841073705c4712ce61a5f6a1f187","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"c9a1edc7ae0bb766dabe801d4929daefb61aa24b50f8c63f38bcb8af792e2228","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"d5be1fb8fe2e82e889f5eaaedac84a4a00ef7c5c6b41b2108eb8540b22b07581","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"fd6599f4c6c55ceeac3c1ae61663f7a5458c909ab12314e9a482a5b6a2634b4c","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"2fb2b7275546aae4cbb5bbf4d1e239f8932b3c728812b0059378c146530a63ed","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"7a840b91061ff0af51b220fe7443708c3535b528f8fe774ff794093ec5db4079","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"aa3d3b78d06e918026ea023b3309802065b9bed031da992bfb45ba123aedd637","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"75b651eabdea9cf05b811fb869e4d0efea60a50a0aeb9059764e17f9e460c9d8","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"db28d042f2b14d0fe388c15b794fe81de0bfd096e68efc45c3bf3e053ec5acb9","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"73bfbc44fad108143e171bbf97a95b5f91eafc179c7d29937a65cfbe9f92f8a1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"c90076d878f9e553a3b8fcfda96a825d28ae20d827b83cc6babd2b231aad0e9a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"26b99350b83fd3730a7422590c59e716e3423a9beeade70178962a3df20e9d2d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"ae9082338f5b65d0da1e517e9e30ab74a35780a2932ee7403687935c78b220fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"cd3152d6475f26ea3857e543253f0170ffe150778a34c8e6ecbc4df584e83bf5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"59d8d27a13fff700aafda8a333016e8f32778752d201dc9609ac6221cf691219","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"a97a8e194bf3a3449ff98f0d29e284fd707c4965c8a0670ff4663ddb32f4122c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"187c4ee253da9cb28c9bbd4801363792e72f686e601a1e47ced24b91614a9e1d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"2afaf285e788c234e54e18f35763b2338b6a5303a0589fbff2a3e3d1f7ce3b77","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"a03d58645884c590f830b3ba7e242bd214d00c41468f29403e05b68d3534c6f7","bytes":5344}},"requestSha256":"14fe82ac98fad6a7ed0ce55e4037d7e62fa4c35de1b641b24ced39586cb48b8a","metricsSha256":"6ecb2f0514a7613de6bd139f6436ebe3cceab5ed00a7702e436450f498e20e30","continuationSha256":"13640b92460533e8dbdd7f23f2c8b9a9f8345d4b5c496fb45b2ed150c5cd5218","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"73bfbc44fad108143e171bbf97a95b5f91eafc179c7d29937a65cfbe9f92f8a1","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"c90076d878f9e553a3b8fcfda96a825d28ae20d827b83cc6babd2b231aad0e9a","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"26b99350b83fd3730a7422590c59e716e3423a9beeade70178962a3df20e9d2d","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"ae9082338f5b65d0da1e517e9e30ab74a35780a2932ee7403687935c78b220fd","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"cd3152d6475f26ea3857e543253f0170ffe150778a34c8e6ecbc4df584e83bf5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"59d8d27a13fff700aafda8a333016e8f32778752d201dc9609ac6221cf691219","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"a97a8e194bf3a3449ff98f0d29e284fd707c4965c8a0670ff4663ddb32f4122c","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"187c4ee253da9cb28c9bbd4801363792e72f686e601a1e47ced24b91614a9e1d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"2afaf285e788c234e54e18f35763b2338b6a5303a0589fbff2a3e3d1f7ce3b77","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001/continuation.pt","sha256":"ecc74922b9ae3e149d9dc410d92d45d3fc632d7e9b54fbb5248289c37a567860","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"55cae209132436c2ebf0712742eff9cf12cc8f43b16de75c683c6419d49c6661","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"08d0b1068a4f96e20ccd32695f9a3452634e75d2b7915b8de80c64b2f079d1e9","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.98--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2487120","recipe_identity":"3b05d8aa02fba09ce8afb971778fb97fdf77d1137b2095bd206cc705379c8889","loss":3.3823749589014698,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/result.json","sha256":"711545c22a115c571ba1e982917c9acfa6b785ad86dc8620e7c76d5b31c3faa2","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.pt","sha256":"0860f01607294cd0e963a95851600cf145beb36996ebc034ffe6a45be362cbe7","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/final.safetensors","sha256":"57cdfdb5c792028486327b3e50e47b47d1a9302dab49ce8be9cace763da0cc83","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"0de2c9abffef43da03efe2fbd0d37bf9175d8a19018bacd0bd6cfabae84ec371","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"735e285bb56f311adb8616c470704248cc407f38708636c7274bfb12df885edb","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"2860d7479c2433789383f9420ba78e2bad7a53a980b310a1cb8f850ab342178a","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"980472c56b0212cfe0c1da51c7c187d6d2d87f55e027aa0837453115ba2db123","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"4ed05095cff24dfa202e25bede1b38d0820612eeebdbb4d2c105931e77a5da2a","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"ddb14289b150ca81dd53dc26f4473b91298e62f623ab89dec84ee62cfb0f4d86","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"63b2cac90dfb15d0445906409f88de6bbca34367a6938e814d13efa751306e79","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"56d5d96f497bedd561b65c9f62c8ee6de24a264d7805feb0d7e1ccf44f2db813","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"67958cf0d829bf44deff8aa036ecae19ae35aaa2ac8e2b5cc7c31682f2e9bf34","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"5686ae856dd79649d9f6021d1726c814e2eda1b2c2274b8c84cbaadb860521a6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"c6b844657d406daf227735e086d49a2f7b7c141fb9d07c54a16b388c5bae8496","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"5fa76c530c03c4d979a22d7ffd23ff1d870dfa1d79a1f50d25a7319e81a97d04","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"c974f5ad295ec0e0b9202cbadde0875220a1282cdc0d5072003a883558d220a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"15d85d8130eb0bea28bbeada94560f2ddc8d7a758c172b61b6ce9ecf13c32ab4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"89cfd69d568268b23e1b6fa4afe43a46f938f5d767def88a3edd5ce4a5609f15","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"4fe709115225b8ece1af20d4e8ac71ef8ad9192a692f6a8a517268f3e1d9f399","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"14ca2ad52cd5a959352fc742e6455b690c4e42324053056426ea43e2274db096","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"fd39280bf432648186b3e2a685f5d8d99ae2d9e62083e5ddfb81bf962913b660","bytes":5308}},"requestSha256":"96111f6188c9ddb50bfc7eabb878bc4f55e6634c59cd52da8c23172dacec3a47","metricsSha256":"a1593190f207d615679812fb3767a9b40a4d81891d29dcb1311f7e34d71e07f0","continuationSha256":"3eebe7be6f46a35ef5d823d4a7d2ff7d87b849ee4d25f7d310cad0d3b1a19ee4","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"67958cf0d829bf44deff8aa036ecae19ae35aaa2ac8e2b5cc7c31682f2e9bf34","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"5686ae856dd79649d9f6021d1726c814e2eda1b2c2274b8c84cbaadb860521a6","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"c6b844657d406daf227735e086d49a2f7b7c141fb9d07c54a16b388c5bae8496","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"5fa76c530c03c4d979a22d7ffd23ff1d870dfa1d79a1f50d25a7319e81a97d04","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"c974f5ad295ec0e0b9202cbadde0875220a1282cdc0d5072003a883558d220a5","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"15d85d8130eb0bea28bbeada94560f2ddc8d7a758c172b61b6ce9ecf13c32ab4","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"89cfd69d568268b23e1b6fa4afe43a46f938f5d767def88a3edd5ce4a5609f15","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"4fe709115225b8ece1af20d4e8ac71ef8ad9192a692f6a8a517268f3e1d9f399","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"14ca2ad52cd5a959352fc742e6455b690c4e42324053056426ea43e2274db096","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001/continuation.pt","sha256":"2c434b9cd1a3a122d56c9efd8e07a981bce021f61be8d9323607b270121fc391","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"264866f8998f2397795ea697994cf8fbd621da67605308f631e1ddd2fda56873","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3b05d8aa02fba09ce8afb971778fb97fdf77d1137b2095bd206cc705379c8889","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.98--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2487121","recipe_identity":"fccd712c18d17f750382ef5a5cb1b4b3dcb4884e39cb0773ee80b11a6ad637be","loss":3.387902159371044,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/result.json","sha256":"4279dbe7a43b1c255e2f1f815200cc7d40f06fd75937b7f78a666c9a2ac7d6d2","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.pt","sha256":"3369d61c18d18c421ccca426a5f59291b42db610185d4e1ae541f731943b82c4","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/final.safetensors","sha256":"6e8dd5b2cf8e0f8137affa75c9a4f8035e63272861940568995cf0c50c7f9535","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"f603e6f8f62266c11c15988102a187d7460e3fead546a13ad416cdcbfcf3943d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"d5251be979f67771f2da9fea0f366b4328240f3cdaf0dc76239358eb7af84fbf","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"620a805612347c02e6afca79b694f18368451f394297561552dc186ffaa84d51","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"6e79a6b369477ed64a84c896a5de2f1aa0933e68e5ef3dc743b9b9a2df75542a","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"7b9124548d743f7ec9ceb457be566852f47d5153e562266d209837c166fbf83d","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"e8ce053ce8fc2203edf8e1b6a4f9542159f6cb068f4607b230fbd3de25c533dc","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"da210dcf81071934a9d63921e2421fc4260a2c9675067d330e89523312683a6a","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"d039ecc1030ac4a2237d19d44b49c527846d8a3a44e43fe0756706392c077eb3","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"b23d82bfd54b140d644d27dec68a7b7f1d40a6f5f2e04f8b2da6afb204bba118","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"3979ffa739a7b54d1f61ab66fa5ad7efaa659cef9cc900ab307db1e01689e582","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"cf55855b3416d24af7f5baf6a99ae750e53f0c3b9dd9e51f4ecc767d18c0cdf2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"9c8f8d7d222c0b04681701e5af20bab64803c65cd78c8acc6331f535f374efd4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"a792ce4712f476fe07494fbf4139e07dcb536e3e55ef7c2101a32f95af55bc04","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"0df88799f7e4d7c65a0d563f8554630956964e721e2ebb53970501ca7111a184","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"2199ea39db8ae978701175638534f1289dabc850aafa7712e066b187893cdfc0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"b405f7a07054132f38f37ae2172000b2d953f30f64193bbcbae05de0f1b69c4d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"8a059aa7c1a667c8c62735544b92e04b102d495d1cbc4612b3ca028270e55d03","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"a0cc59c9ea39035bee50b0b87b82f08bbded4abc60364cb7fbc60116a39b48f7","bytes":5315}},"requestSha256":"b1c7e83701ed1f1fda624dc62ddd5068604723f78d13dbd7b76eed5a16efec97","metricsSha256":"bd817a99add7de8330c742a1e50270de10c3bb6a0325461ee2202448f9d900cd","continuationSha256":"04b2d98680b5842345779afb6b3cc34e3ee017e4c21e146dd9e8576f08917682","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"b23d82bfd54b140d644d27dec68a7b7f1d40a6f5f2e04f8b2da6afb204bba118","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"3979ffa739a7b54d1f61ab66fa5ad7efaa659cef9cc900ab307db1e01689e582","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"cf55855b3416d24af7f5baf6a99ae750e53f0c3b9dd9e51f4ecc767d18c0cdf2","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"9c8f8d7d222c0b04681701e5af20bab64803c65cd78c8acc6331f535f374efd4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"a792ce4712f476fe07494fbf4139e07dcb536e3e55ef7c2101a32f95af55bc04","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"0df88799f7e4d7c65a0d563f8554630956964e721e2ebb53970501ca7111a184","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"2199ea39db8ae978701175638534f1289dabc850aafa7712e066b187893cdfc0","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"b405f7a07054132f38f37ae2172000b2d953f30f64193bbcbae05de0f1b69c4d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"8a059aa7c1a667c8c62735544b92e04b102d495d1cbc4612b3ca028270e55d03","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001/continuation.pt","sha256":"43d0dcc40afaf1cb3b603102a9d2cfbd27c372677eb28441a4e43aead75d2d5f","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"05376aed83bf802b53f7cd41f3660d9a87be4c3e2c01855c72bc02933444dc73","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"fccd712c18d17f750382ef5a5cb1b4b3dcb4884e39cb0773ee80b11a6ad637be","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.98--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.98-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2487122","recipe_identity":"56137731fc5d677f4412f95592cdb7f6328a7e1cf6bbceb1d13d30494bab802a","loss":3.3818383573839106,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/result.json","sha256":"3234d5c2560e771e6aae88d071930546f8e494a0a3ebf768e39998e11e992dc3","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.pt","sha256":"c6600039994bae766736ce64008262c201d502d2c80ff196625b1fd0fc31e178","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/final.safetensors","sha256":"02c8d3c1d782744a82be141d29001f2bbb28eab6423235e3431d27489d63514d","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"f199c1ab0467c2de4b05a190b0607db20991d25f0ab107bf9b7e5a0d9b97288d","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"35b2b146895c865ad13c91ecaf5ef1eb34a6dda28e58f9797294c9261239cb18","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"7550ffdb932ce4fde1809202494adc5bf7a98d343f824b052d13c6a5a3408383","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"6274b42a8851325f9cde9eb11a39d4d1296a188cfc02e0c4267cc7358e2b755e","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"33fcce37fb314c17aff726047d4029ef60d54fb99000e74ac7cb35b9969ab675","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"4052a9c9edcb0630232b93709ff91e2aee6491389a4d2100ee53f895d73f94b7","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"194b90e537cf0ac3c6c01d9e905469e70be5b7f5ffd3f09165bff4aef3c2a0f7","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"38330a75ca793cf55510e2eec092e586a4efe0a849df4636b70c50c242089be8","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"65b47a0f8d24f5218b91cf8e28bf7548487253d4fcaf04d59bfe3f39ee370c63","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"9a447f15a7d7e8ddd63a560d06537cc10c86847db5c6c7c277725b7f70746a3e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"19f8fb95a08616eb61f3941c79ba1ea99be9ba8d3bba2325cd14cab64233a2df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"872213dc6c6fbff5ea21293e6fe22b2a03ed45b01d9188ca8b120f54cfd4da8b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"7deca9dfd0c6c434be42e75c30375c56cd023cdf2f6ccd4a6773b77c190db84e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"55c87869edbbfd68c777979793ccad6717a7819e7e688f7591b6a4f4d9d3b382","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"e40d0ea4fc93641c915a0205ae5adeb9a5b998cd951d37551cf34ee187579571","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"5f00173e341baeba254d568ed5dab3e16771cf1ba817e3469bfa99eb7d2e233d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"9cd7083d0ccca26fc5408792f6c1c546f0f470ef23157c2454c43daa5f7ae08d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"3a64d51ef2243d8b7814232f044893fd8da1d716f4f3c413f7aca250b0175214","bytes":5343}},"requestSha256":"8499ab08bf749e95763f0d70a6635a2c934aa74620204d26f58224268564b204","metricsSha256":"8bf96b4923a5483538f4a422c0f90f2e0771ab62fe1c1d8eb71b5c82a85b85a3","continuationSha256":"29ecc2222fb331a2c8c6799038483cc8095814983cf06c66d956fb6307429aa0","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"65b47a0f8d24f5218b91cf8e28bf7548487253d4fcaf04d59bfe3f39ee370c63","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"9a447f15a7d7e8ddd63a560d06537cc10c86847db5c6c7c277725b7f70746a3e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"19f8fb95a08616eb61f3941c79ba1ea99be9ba8d3bba2325cd14cab64233a2df","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"872213dc6c6fbff5ea21293e6fe22b2a03ed45b01d9188ca8b120f54cfd4da8b","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"7deca9dfd0c6c434be42e75c30375c56cd023cdf2f6ccd4a6773b77c190db84e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"55c87869edbbfd68c777979793ccad6717a7819e7e688f7591b6a4f4d9d3b382","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"e40d0ea4fc93641c915a0205ae5adeb9a5b998cd951d37551cf34ee187579571","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"5f00173e341baeba254d568ed5dab3e16771cf1ba817e3469bfa99eb7d2e233d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"9cd7083d0ccca26fc5408792f6c1c546f0f470ef23157c2454c43daa5f7ae08d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001/continuation.pt","sha256":"b507912c4ffd740fabe25fa4356f40e71717610bba7a359c568e0ff3bd441490","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"6a48c879e3c73bd0914fc1aacc6b292a16299252e07e96cbf0019d222e7ac1df","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.98,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.98-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"56137731fc5d677f4412f95592cdb7f6328a7e1cf6bbceb1d13d30494bab802a","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s42","horizon":160,"attempt":1,"job_id":"2487123","recipe_identity":"3d66f3cda284dd562a274a6f3ebe5b5a73b8ca577fb5e34320ecb6149147ea40","loss":3.3282830529138687,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/result.json","sha256":"126f758f39164d2cfc279bbb1b822f94acb9dfe83843a64ef1b2dad94202aa88","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.pt","sha256":"9d31fbc6e0318d0e23845b16978a33a6bf59b4197c0d58938b9e3c60fce5c98e","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/final.safetensors","sha256":"0068b97fa6e832365e44939345784e1eb6d8271d03a797ca08fe892647a27c88","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/final.safetensors","sha256":"1e94e108f404ed897b56c1ce2006bb6d1dd7aa62479b66778e385a32926c9099","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/final.safetensors","sha256":"ac9a6fcef844b49c0d49db870d0804d7ee3b5885d4a115d2ee1f57e87a1480f0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/final.safetensors","sha256":"6fe3808beb5ece78d5ed9c682dd4e8f8e3a703d19814b0956b99612d1ed962e2","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/final.safetensors","sha256":"b85fda3fad318820588877d05fb0617448ec3b56188114349a5b8caef8f86836","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/final.safetensors","sha256":"0cf1e3058ee2ed401329fa65c8aeb75e97d281e208c966a23deed6a1962f41bd","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/final.safetensors","sha256":"bf17c46a38eed785ae4b3abbb149bee31a4386fe78d3349be487b000d6cb875c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/final.safetensors","sha256":"1cb0ed4450a9f66c36e5f48c51d9308d6a86e88a747984dbb2ea46125ea31f5b","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/final.safetensors","sha256":"fc5b46149c891d4f69163d2943ba6fdd8d1bb42fcb4a469d5e67f0c1bb669462","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"05a19adeb45a616985a272a6519861fd8093f33d2b407803241f4d3fafb3ab59","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"384cf9c17288168a858ed8523df55aa6ca2871e2d978a38fb2910be30c19c235","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"eea86e38756b4a88e29a2f92369e40ae7b4975377fc1efb2eb08d1ceb50450b3","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"2645edff0bfaf454271b9a0e393f3df1758fbd12e06e56c2ba904166ad0d93d4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"a8aefe522d89d1db90ba0acf641199b641f616a8901e2e2a9e12c1b0a0b8481e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"36dc67ac3f50de46550b343bbdd3b0bcc43cd81badde0b2625f8bcc7ebed76bb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"25fbe1855dbbc4f4cab79c9940c8994b4d17af907401fe1a21299a4ec8dcc3be","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"c09575bd7dbce56bbd2b6516c83684917ea8774dcbdf622800c2f9fba4933e44","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"e4f85bf9b9e1dc56bc78a18620befd72725bb5897b4c953ce17649ee10b3ab90","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/environment-resume.json","sha256":"16fb334fb057b61735ab437d01a308584a956d4e2a6431bff652ca26e99f256d","bytes":5254}},"requestSha256":"8069557e9b87caf89eda8295bec3f84bedcf203aafe4dd5e51ce60d524bf0eae","metricsSha256":"fb4d7c4fccfcec8afe862fc59ef27b330295f4c5a6a0beefee83c1322baa9f4a","continuationSha256":"9ffadd5afc926d2a1a86a2098af3bf464fd8d22108f3ef40c2b961f6b8607dee","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/epoch-216.pt","sha256":"05a19adeb45a616985a272a6519861fd8093f33d2b407803241f4d3fafb3ab59","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-000/epoch-216.pt","sha256":"384cf9c17288168a858ed8523df55aa6ca2871e2d978a38fb2910be30c19c235","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-001/epoch-216.pt","sha256":"eea86e38756b4a88e29a2f92369e40ae7b4975377fc1efb2eb08d1ceb50450b3","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-002/epoch-216.pt","sha256":"2645edff0bfaf454271b9a0e393f3df1758fbd12e06e56c2ba904166ad0d93d4","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-003/epoch-216.pt","sha256":"a8aefe522d89d1db90ba0acf641199b641f616a8901e2e2a9e12c1b0a0b8481e","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-004/epoch-216.pt","sha256":"36dc67ac3f50de46550b343bbdd3b0bcc43cd81badde0b2625f8bcc7ebed76bb","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-005/epoch-216.pt","sha256":"25fbe1855dbbc4f4cab79c9940c8994b4d17af907401fe1a21299a4ec8dcc3be","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-006/epoch-216.pt","sha256":"c09575bd7dbce56bbd2b6516c83684917ea8774dcbdf622800c2f9fba4933e44","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/node-007/epoch-216.pt","sha256":"e4f85bf9b9e1dc56bc78a18620befd72725bb5897b4c953ce17649ee10b3ab90","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001/continuation.pt","sha256":"a979576869f0a17eeddf2f6888a30d1b9a9d5d50e252650df61aa458befc1dc9","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"8d056b802156d1eef5ef9ee9728ac889d0e2f2f724fe77f7990084c25263c4f7","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"3d66f3cda284dd562a274a6f3ebe5b5a73b8ca577fb5e34320ecb6149147ea40","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s43","horizon":160,"attempt":1,"job_id":"2487124","recipe_identity":"bcc51bb1a622b69ce26e7be417b5285f21f9bc1d463a5b44b35a392565bbab0d","loss":3.3353299228975146,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/result.json","sha256":"0c6fe5f5bd8fb1ef9d935e9ce2a166a0d27ddbdd5ed32a91a146ce9a6eca418f","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.pt","sha256":"b09b197d06fe57240e88c2aa37cb7f86bd06554b736ac03f2440d3e457dbac5e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/final.safetensors","sha256":"6b895b152ff2c7df588d5330bf0723c4cc6e26efb8cde878be3bfff05416bb35","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/final.safetensors","sha256":"dfbe4976f34fc347706a7db650e5e687ba295628bd459104f1c37fb284707183","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/final.safetensors","sha256":"915825561a135c7e470b5a6b2353c2b9416cf581fb216d5e4677c3a1018323cd","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/final.safetensors","sha256":"c23129aa3e3d96d18b781320cc722b93bf065c71c69a939f71a1b134e56b58fc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/final.safetensors","sha256":"6e0bd748083f1eeca28a59b9a74cf6810ff30b90612ed5d33d266c4b4afb4c95","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/final.safetensors","sha256":"f7e98af266519008999f3c43dd77adcf78ff12938cb273ec0a3e9f2ffd913bc0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/final.safetensors","sha256":"290c7e1bf85d0b3940dbeae0ce38cb107f94f06d2c34835eed7e67dbc07eb427","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/final.safetensors","sha256":"abccb1f7c0830dee06ab4fe12d834d6d3dfb8296ded1693de4cd97a03ef2eba8","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/final.safetensors","sha256":"974fd9a9128ef1bfa13fb8c34aa9fce597dae2d3be345cee0777e479b54d214e","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"7b0eb71271aa852af4024c1fd577588c656d8e0f2eeb76bc2df82c2dea041d5d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"896b5fde1f4cdc96c346a25ccbea39c9de3792e709b615a065a2a5c176505a51","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"dde3ab2f345ef0a5068e543086bd539d0634024273d796d7474196f288813727","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"49b392e12e761984965e274f600f67ec424a83b3888c5e03c7b33f3120ca2102","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"ba7fd3717320feff954f2e24d3a3d2a637ed87d4a63037eed96914746298f414","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"811c162e26500d5c175553842377ef98030430bd0dbcf89fb160654f44841470","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"acc824203032115b487882493da295d749895fb65fcaa84f05e74d75bc6a4564","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"ab6d6d36330e2b388910fb514d15dce87fadb40d99431e6f306f0686393b927e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"f75ba20242c596dc9eb23edd4cbfc42a980d443dec1e90f6091743d597dc1c80","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/environment-resume.json","sha256":"fabf688558fccf23f92e4a27ae663c60c6ff1b1a42561d46b4e68d80f7a97d6b","bytes":5344}},"requestSha256":"3e05b226e2b12357e328d291018030afa29e6538f5b0cc76bade58e9072d2b28","metricsSha256":"b7d37aa2ba73a4c97fd4b66000387a582ab3f318025b25fababbb633a4f5b8f4","continuationSha256":"46299594eb94d6ead77b8359726bd82e17f59da6b1e90e656afbaba4fed535c8","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/epoch-216.pt","sha256":"7b0eb71271aa852af4024c1fd577588c656d8e0f2eeb76bc2df82c2dea041d5d","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-000/epoch-216.pt","sha256":"896b5fde1f4cdc96c346a25ccbea39c9de3792e709b615a065a2a5c176505a51","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-001/epoch-216.pt","sha256":"dde3ab2f345ef0a5068e543086bd539d0634024273d796d7474196f288813727","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-002/epoch-216.pt","sha256":"49b392e12e761984965e274f600f67ec424a83b3888c5e03c7b33f3120ca2102","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-003/epoch-216.pt","sha256":"ba7fd3717320feff954f2e24d3a3d2a637ed87d4a63037eed96914746298f414","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-004/epoch-216.pt","sha256":"811c162e26500d5c175553842377ef98030430bd0dbcf89fb160654f44841470","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-005/epoch-216.pt","sha256":"acc824203032115b487882493da295d749895fb65fcaa84f05e74d75bc6a4564","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-006/epoch-216.pt","sha256":"ab6d6d36330e2b388910fb514d15dce87fadb40d99431e6f306f0686393b927e","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/node-007/epoch-216.pt","sha256":"f75ba20242c596dc9eb23edd4cbfc42a980d443dec1e90f6091743d597dc1c80","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001/continuation.pt","sha256":"6358451745d0eda605b5064ee86c27889b2e6bd3e479b2e64b8d54768f9cbf21","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"0b86a5f6d8887085fc19a165db95b23f154d99f32e80d2f44a2ba97a063f037b","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"bcc51bb1a622b69ce26e7be417b5285f21f9bc1d463a5b44b35a392565bbab0d","loader.blocks":3188096}}},"awc8-wsd-h160--0.003--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.003-s44","horizon":160,"attempt":1,"job_id":"2487125","recipe_identity":"f6e67fae44fe327c259e8ed4628dafde40b0fbcc8684938824caa5d0b30f78a6","loss":3.3331716778774796,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/result.json","sha256":"e049ff6f6055ab45278f2cd05dfd96ed05497f690f75a7646c1940f4c6303639","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.pt","sha256":"a500439ce674829a9c2b56dabc621f48848beac9d322ba926f300e881e7ac54e","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/final.safetensors","sha256":"6a3882fb08ade1018146586876f36b3719df9fa2f49173143cf8aceff63eb4df","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/final.safetensors","sha256":"19c29b393ddb68b3463830f2cedcfa0bf73b48c0cf1d71051d1408c50c7d56b7","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/final.safetensors","sha256":"c4eb52a9473dd477483592e4f35ccd812cd361a5bb4442db791aec62f7f0550b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/final.safetensors","sha256":"39c8c2b8c0b05b856b3c76d86935fa7f80ed9d531b5fe2217466572c7822ccdc","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/final.safetensors","sha256":"35d53b650bc9a3bf3ac2a3833b7e0e5be4713924c9f8c02e22f57815e419d7b3","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/final.safetensors","sha256":"820a92fae89aaeddbb49e22d4c8519b072dc7830de4a1a10afbde511ceb3f641","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/final.safetensors","sha256":"1b5b39d89952cca1665972f310ce078d812207d72ef70014c8fbf4366a527926","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/final.safetensors","sha256":"644f245ae11841625996eb9373ff91c9b4f2e6e0acba1b145aa5530ee9067fce","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/final.safetensors","sha256":"5e21ce82c2ce2e16c2161eb6357a671698253fc4b1e140b16078a03586e7d127","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"84ebccfd27b44aafa8507e0b66f59f150095616c003decfd6bf7cae1277172b9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"067ee25d80dce401c084b3d04b306e38ebf10d413dbeda11536962649f9a81bb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"76767bcf8e38b0a617798d0f0ea32fdc08b206f3fedd48787473a0f31f496dd0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"266305d54c6bc6906dcc12be2d012cca9281055ff9273fbf1635c85f9f42022f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"b09ce243e78acc277f888d71bbbf1b64a0a7dd5dcd23cb6575e344aac9fc1167","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"961004658121086e78f84dc0353e411992a7c4d27ffaa7b222a85c1e2d21572b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"ae0a193f02cdf4aaad41abb753f2d8656d6953604378225da88b1c0e8b76b3a7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"8edc5eb652abfa9544578c52a1358f23430a2587a4713a70e742d8bc673ecf25","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"7399f44b0a5b6ffe130f8babb6ffc10df3f6b03af16d0c354fa300cf822c3f50","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/environment-resume.json","sha256":"65424f3d645b22b0aa411df2379b13ae27ba7f97ea8cf6066114b9cf55749545","bytes":5257}},"requestSha256":"38040f48340c5457e97048a12ede27e14a8f3928317b4529e4675db116551641","metricsSha256":"879fade9ece6197855c52de5210aa86db8d640eb4ad7ba100efa158225f5527c","continuationSha256":"7d343c984bf0dacd629f6abda0fd4f718af9202eb099dacfd5c4a743d3dffed6","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/epoch-216.pt","sha256":"84ebccfd27b44aafa8507e0b66f59f150095616c003decfd6bf7cae1277172b9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-000/epoch-216.pt","sha256":"067ee25d80dce401c084b3d04b306e38ebf10d413dbeda11536962649f9a81bb","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-001/epoch-216.pt","sha256":"76767bcf8e38b0a617798d0f0ea32fdc08b206f3fedd48787473a0f31f496dd0","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-002/epoch-216.pt","sha256":"266305d54c6bc6906dcc12be2d012cca9281055ff9273fbf1635c85f9f42022f","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-003/epoch-216.pt","sha256":"b09ce243e78acc277f888d71bbbf1b64a0a7dd5dcd23cb6575e344aac9fc1167","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-004/epoch-216.pt","sha256":"961004658121086e78f84dc0353e411992a7c4d27ffaa7b222a85c1e2d21572b","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-005/epoch-216.pt","sha256":"ae0a193f02cdf4aaad41abb753f2d8656d6953604378225da88b1c0e8b76b3a7","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-006/epoch-216.pt","sha256":"8edc5eb652abfa9544578c52a1358f23430a2587a4713a70e742d8bc673ecf25","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/node-007/epoch-216.pt","sha256":"7399f44b0a5b6ffe130f8babb6ffc10df3f6b03af16d0c354fa300cf822c3f50","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001/continuation.pt","sha256":"6eeed4caeab3d3d1f62c5c2eb390fa566f812d93d75b3e52c4c92ca46d953404","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"d8a416a40fcda3f5941119ba37864b1b5c91bd70344855dac9a2d9a5a1b09589","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.003,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.003-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"f6e67fae44fe327c259e8ed4628dafde40b0fbcc8684938824caa5d0b30f78a6","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.999--s42":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s42","horizon":160,"attempt":1,"job_id":"2487127","recipe_identity":"4a6d0463909b44a3f9a51875c5336c1e2b2e3afd1cce535c728f2e412ddabab1","loss":3.3434955812164113,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/result.json","sha256":"9766e8a6be1f519b99a22c13c100969e1efbefc31fd440444929d368a9e8e3cd","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.pt","sha256":"363edacb3541efd46deef623e52e9c6628f28d1b3a7b5ac81a2a3098ead95e98","bytes":1958929429},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/final.safetensors","sha256":"41dfbe05f79cfedec7ca15dc2ec9ff1fbdd1476cc65d909d525ba6caf14848d8","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/final.safetensors","sha256":"e82ec20ba9d623f11e8eceb09d3d8ae129343dcae65ff3a46ebc5e52ff67e4e5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/final.safetensors","sha256":"81ded93ce5c28c25a22cce469f12901b5924be65f0fd598a5b28e2105400eacf","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/final.safetensors","sha256":"03c5cf6c6250c66a8893ff6f008df30757a21c6a564914ddc4a97ece635fed9d","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/final.safetensors","sha256":"5464b1277d308a78f81adffd9401cbc7a0a90de303c4d0358d34082129351207","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/final.safetensors","sha256":"2c21efb42dc7379ab5dfbd979f57fce3bbed91d9c2a1f93308190c8bfce1d0e6","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/final.safetensors","sha256":"eba0430b8b768b86174f69958fc6be7f3c7a3e0372f729df959692ca0395812a","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/final.safetensors","sha256":"e21dc5b49fd7fa500261822fdfd495b1254c17fe58630f174d4b4cf311ac0b1c","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/final.safetensors","sha256":"0def8616527bad16ed7124a90be45aed1997eb4c45812344d0413e5a7e97955c","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"a0d8231e33e2fc09daf90bef695709e7b758da50b0bf643de348f8bae04e1d9b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"8b73148c05b2e544e4abaf5cc3e9e46fa29f4ec5f26cf78be18872c3af55c51e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"d917286c9dbee3bdfceea821e82024120fea79ab8c24205f99863279b7dd250c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"454797fdf04d111638140de9bf3f3f3b6f7e64555b2650fb35645197afd9e1b0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"3d97728a3e62b8fc222d959606f416d94d2a898c3e824cdd42fae953fa3e1fbc","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"9e4c240e0a3463892ede3b8142e488ddcb46108e4cc226541f8bacb35aa4b4ca","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"9cebc6fc98b7707275bd5b358d91426a7912e4ad67572a3396a005c06aa83683","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"3b4dccd4c989d6a093979fc6de41fb3aac529723afea2e1d02f26ad0e2b0888d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"9f7accd715cd2ab549d47a0e0b96fd7be9a8a4e777c80b36f220a8cdc1a015be","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/environment-resume.json","sha256":"e624727223f03ace00aaa93f67ed9698908c301921fb01d021335fae78459830","bytes":5315}},"requestSha256":"176a6a455a71893006b68cef5421edeb6a6c2017e498f53fd747339299f1d965","metricsSha256":"4fadb09f40c9b2fb5325c3ea33e42d4c87f9a984df2726f5865de70059111740","continuationSha256":"d6e1f8d393521d6e4e53f3da138c62f9dced498c67879fd0fccf09a8d3aebd1f","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/epoch-216.pt","sha256":"a0d8231e33e2fc09daf90bef695709e7b758da50b0bf643de348f8bae04e1d9b","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-000/epoch-216.pt","sha256":"8b73148c05b2e544e4abaf5cc3e9e46fa29f4ec5f26cf78be18872c3af55c51e","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-001/epoch-216.pt","sha256":"d917286c9dbee3bdfceea821e82024120fea79ab8c24205f99863279b7dd250c","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-002/epoch-216.pt","sha256":"454797fdf04d111638140de9bf3f3f3b6f7e64555b2650fb35645197afd9e1b0","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-003/epoch-216.pt","sha256":"3d97728a3e62b8fc222d959606f416d94d2a898c3e824cdd42fae953fa3e1fbc","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-004/epoch-216.pt","sha256":"9e4c240e0a3463892ede3b8142e488ddcb46108e4cc226541f8bacb35aa4b4ca","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-005/epoch-216.pt","sha256":"9cebc6fc98b7707275bd5b358d91426a7912e4ad67572a3396a005c06aa83683","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-006/epoch-216.pt","sha256":"3b4dccd4c989d6a093979fc6de41fb3aac529723afea2e1d02f26ad0e2b0888d","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/node-007/epoch-216.pt","sha256":"9f7accd715cd2ab549d47a0e0b96fd7be9a8a4e777c80b36f220a8cdc1a015be","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001/continuation.pt","sha256":"2e8be6a8723ea1ed4e935a9d37788df1130c72e2fffd46233c4d42bfbecd45a9","bytes":1958918101},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"b66b61f02e00729b78d54de3e5614281e0401911f630f4c0d280d701c3e4f956","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":42,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s42/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"4a6d0463909b44a3f9a51875c5336c1e2b2e3afd1cce535c728f2e412ddabab1","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.999--s43":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s43","horizon":160,"attempt":1,"job_id":"2487128","recipe_identity":"a73689e57dda16885878c6f30b1bd4b243416b95677169b76c25ab0a6f576417","loss":3.3479205570656556,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/result.json","sha256":"20f527985751f0574f6ffc3e20d3956bc5cf3ffab479ed8398f3f0ad23ee78cc","bytes":890},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.pt","sha256":"4730a81a0e161655a1c62be84789ae85add5b80c527b03d121d75394ef3e6662","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/final.safetensors","sha256":"f0e3b607bd606645854147368d0c352d5e4588a11c66d228f3fe92d068fdcca5","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/final.safetensors","sha256":"5efbf8de8a3aca97fb6f6eb024ab6f8f163aeb1a031a5185389a6c013fb4cbb5","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/final.safetensors","sha256":"4252f93fd5842acdcb7234b6d8a1350d3506704012cf9b98d352022477a2a5d0","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/final.safetensors","sha256":"38fe277f8f6cb7043c519bdf08bec63538663c6e8ce7a5ab28db4a31f29a12c9","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/final.safetensors","sha256":"984e57a385328775270096ffad7109b79bf19c84ca5148bdad8c3be6e4261ab5","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/final.safetensors","sha256":"68c717704f4ca39c93139dd4c24afcec39cbe98b6c6c9376552b5f612aca7866","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/final.safetensors","sha256":"5576703df34d42514c4a1a024b8cd84609b35619601d0300dcde8b676bc62cce","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/final.safetensors","sha256":"58f8b85faa26415b055f5c7d52d9da3d85f37270778a90eace41152bf2daa29d","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/final.safetensors","sha256":"afa45049e1242f9c60ccdea5af0e925127f092c911cd801842d46f9027236b1f","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"9fe148fe88e88f484d6646d5a8714d30ea44c97ad581b92571f8f3aee63181bf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"35caaea33f67e8b7f0ee02594ab6418b5cf523f1d1f62d9bb527adf8970049f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"ee452db3d4f91359eaeaaa0cf054b524fc9b70c6212a1d9a0b977daf1b0fc728","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"1d9daee43b2b59a4238725de54f3edf3263b74b3a966ecab63084ff17001f1f5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"f73a2c9fee445c13110d6c71ee525ea83134109e0ae79919542c1bd445372e2f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"cb099f44ffcb8627d9a5c0faba4702779f8ac84fbfe188ddaff8ff4bc78f4e82","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"91f3f486e18995d159596eefbec412cd884cb10339e137ee32ff20f1b98ef968","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"40984f1f037912c51f209bf9f43625896e4359b4111290c58fd162cddcf28ab5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"6f5c35ce2febc184c4221c90d3df130bc46466251cef8627c5b9b570e5f22c8c","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/environment-resume.json","sha256":"805331b409f4e7f6e65349f9c1c94910fc457679d41624e046f0f6b6b3d58d9d","bytes":5314}},"requestSha256":"4031320d014c04372c1de978f3dffe9ad7353133806ec585d901fa3c8cad1c04","metricsSha256":"8146ca43a75ff3a86ea0eb3aa82f7c583341501f1c0a77551c070640d64dc28a","continuationSha256":"451d2553dbb83714ade71fb34eeac80eae5b3232081d555afdade500feacc06a","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/epoch-216.pt","sha256":"9fe148fe88e88f484d6646d5a8714d30ea44c97ad581b92571f8f3aee63181bf","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-000/epoch-216.pt","sha256":"35caaea33f67e8b7f0ee02594ab6418b5cf523f1d1f62d9bb527adf8970049f8","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-001/epoch-216.pt","sha256":"ee452db3d4f91359eaeaaa0cf054b524fc9b70c6212a1d9a0b977daf1b0fc728","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-002/epoch-216.pt","sha256":"1d9daee43b2b59a4238725de54f3edf3263b74b3a966ecab63084ff17001f1f5","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-003/epoch-216.pt","sha256":"f73a2c9fee445c13110d6c71ee525ea83134109e0ae79919542c1bd445372e2f","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-004/epoch-216.pt","sha256":"cb099f44ffcb8627d9a5c0faba4702779f8ac84fbfe188ddaff8ff4bc78f4e82","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-005/epoch-216.pt","sha256":"91f3f486e18995d159596eefbec412cd884cb10339e137ee32ff20f1b98ef968","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-006/epoch-216.pt","sha256":"40984f1f037912c51f209bf9f43625896e4359b4111290c58fd162cddcf28ab5","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/node-007/epoch-216.pt","sha256":"6f5c35ce2febc184c4221c90d3df130bc46466251cef8627c5b9b570e5f22c8c","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001/continuation.pt","sha256":"d97b2867a78cca85f06e8fa683487d4d565513e02a5c541d55dce9bde5ddfb14","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"9c2f048e5f5b28c44e993287fbdd7bf42efd10dbd25d932c0eccb60e5454f1bb","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":43,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s43/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a73689e57dda16885878c6f30b1bd4b243416b95677169b76c25ab0a6f576417","loader.blocks":3188096}}},"awc8-wsd-h160--0.002--0.95--0.999--s44":{"receipt":{"campaign_identity":"3487e1c9649c0e1036887d8d39fe5ad9d5b1a832572de8cb1a2135fd6270bed2","run_id":"b1-0.95-b2-0.999-lr-0.002-s44","horizon":160,"attempt":1,"job_id":"2487129","recipe_identity":"a60fbe6402aa0fa1656a4265d0310d0dcb0fb787524f657bea5e45a6c1ad99c6","loss":3.3417755724939053,"result":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/result.json","sha256":"cb7325e2a0a7fba764a89a097279971076d85b5c99104de747e9d780fd87b5fb","bytes":889},"final_state":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.pt","sha256":"6fcdc4fcc2b759b752e1657bc0fb37e5fcdf1a1c6d95150b5a1289d9f89ecb5d","bytes":1958929365},"final_weights":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/final.safetensors","sha256":"70877151ef7d5937079a33e2f9d47ab91baba5f7b0b6be4ebd0413a6c7ccbd5c","bytes":81621488,"aggregation":"arithmetic-parameter-average","num_workers":8,"workers":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/final.safetensors","sha256":"c5246db770590469b5d81dad79bfcd292598717743f53a77b07b0d5789011b6a","bytes":81621488},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/final.safetensors","sha256":"dcd170f04ae1290bee5122a52249fb3a2327bd439586c79b183fd28736947b3b","bytes":81621488},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/final.safetensors","sha256":"cf316b51f8eadddd1612e62cb130c6c8e25fb234cffa219c1b0d7e769276f133","bytes":81621488},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/final.safetensors","sha256":"a35fff3def1e0fc210e7f9e5b6d76c9ba984259ca17df12725b5c992a824cbbe","bytes":81621488},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/final.safetensors","sha256":"21362d478985142664475fea67d9bf0f671ea5cd775a8dc76d67b0bc69ed67a0","bytes":81621488},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/final.safetensors","sha256":"9664d250ba102c7b49fc7078e099faef87f49ad63c47e547b808d0ca0accbb9c","bytes":81621488},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/final.safetensors","sha256":"f6b86b50b7bd63f6cb81b82ad03267e811f159b4a28d5ab307038fa79bcd7017","bytes":81621488},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/final.safetensors","sha256":"2018b7353a5ac6351c4504c89614f5f1309cf564d9d21f66bd8910f4532ee733","bytes":81621488}]},"pre_checkpoint":null,"parent":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"b1fba519a038bc6f02ac844ab614f170bc8c849898a81921f6fd0ecbef53c9d9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"ceb426ab8196e0e7f9703eac54f40dcb08ea015af161e2440f04094d860447db","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"d3d67c21569f7bd18552ce9d89ca252a3a8f0c612b8749c90a5c311597c74292","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"8375c9964076fda4f573d9535dee2b9c0a93a2bb2b54c73879bb98dd17c1bb80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"bdc5e91f1b74b61d4303ea7f7a9ddb04e44493bda61c331f0bb9ad51fbf5fb42","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"24eaa0263439edb8a9fcee72bf22bc5c820f36dd00cb4009b541150ba6707a7e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"a9011712e16b9a34e8d405626e3b68b0d5b38430e7c547d5df66f98df37a407a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"97a4784f7837868ba0daaed00dff1025581537299429adbb9869bb3321020993","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"c2e8a0ca8cfb4008de1a209eedb7a557e974ce1c556e10ad0f7ba83d6330df6d","bytes":244926383}]},"resume":null,"environment":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/environment-resume.json","sha256":"cfc2cbec8711b782c93d49b3c36bcbfcef253e4ff9547c79b994c26adb74557f","bytes":5258}},"requestSha256":"20c3f5d3824c18d868a2377ee33673d34b92893faf3e568f561ab82961815c77","metricsSha256":"e364b9fcbde5aef44d21fb5adb0669528dc9afbd2ac58d0a2f3d71ad7132895a","continuationSha256":"58cc73c85a5c178eec7e654b0eb58af279b22a6f6bb0ad92f7d5b68e2aba0839","continuation":{"source":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/epoch-216.pt","sha256":"b1fba519a038bc6f02ac844ab614f170bc8c849898a81921f6fd0ecbef53c9d9","bytes":17957,"members":[{"worker":0,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-000/epoch-216.pt","sha256":"ceb426ab8196e0e7f9703eac54f40dcb08ea015af161e2440f04094d860447db","bytes":244926383},{"worker":1,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-001/epoch-216.pt","sha256":"d3d67c21569f7bd18552ce9d89ca252a3a8f0c612b8749c90a5c311597c74292","bytes":244926383},{"worker":2,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-002/epoch-216.pt","sha256":"8375c9964076fda4f573d9535dee2b9c0a93a2bb2b54c73879bb98dd17c1bb80","bytes":244926383},{"worker":3,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-003/epoch-216.pt","sha256":"bdc5e91f1b74b61d4303ea7f7a9ddb04e44493bda61c331f0bb9ad51fbf5fb42","bytes":244926383},{"worker":4,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-004/epoch-216.pt","sha256":"24eaa0263439edb8a9fcee72bf22bc5c820f36dd00cb4009b541150ba6707a7e","bytes":244926383},{"worker":5,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-005/epoch-216.pt","sha256":"a9011712e16b9a34e8d405626e3b68b0d5b38430e7c547d5df66f98df37a407a","bytes":244926383},{"worker":6,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-006/epoch-216.pt","sha256":"97a4784f7837868ba0daaed00dff1025581537299429adbb9869bb3321020993","bytes":244926383},{"worker":7,"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/node-007/epoch-216.pt","sha256":"c2e8a0ca8cfb4008de1a209eedb7a557e974ce1c556e10ad0f7ba83d6330df6d","bytes":244926383}]},"continuation":{"path":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001/continuation.pt","sha256":"0f0ca0ca95ce46bc62f2f3cc73739a1878556fdd24074fde1ea54531d88a7bf6","bytes":1958918037},"source_horizon":120.0,"target_horizon":160.0,"epoch":216,"step":16812,"cursor":2151936,"num_workers":8,"topology":"one_peer_exponential","scheme":"awc","preserved":["model","optimizer","rng","cursor","step","completed_epochs","best_loss","best_epoch","committed loader cursor"],"previous":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":120.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"explicit","checkpoint_epochs":[216],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-120/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"e27c90e1e90959565f648d41149532628af523873b645d237789a82bba9aa15f","loader.blocks":2391040},"updated":{"config":{"model":{"vocab_size":32000,"layers":8,"width":320,"heads":5,"ffn_width":896,"context_length":1024,"rope_theta":10000.0,"norm_eps":1e-05,"init_std":0.02},"data":{"cache_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/data/c4","dataset":"allenai/c4","revision":"1588ec454efa1a09f29cd18ddd04fe05fc8653a2","tokenizer":"TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T","tokenizer_revision":"59f6f375b26bde864a6ca194a9a3044570490064","shuffle_seed":42,"shuffle_buffer":100000,"shard_tokens":33554432,"tokenize_batch_size":256,"prepare_workers":8,"shuffle_group_size":2,"prefetch":true,"prepare_train_tokens":null,"prepare_validation_tokens":null},"optimizer":{"lr":0.002,"beta1":0.95,"beta2":0.999,"eps":1e-08,"weight_decay":0.1,"grad_clip":1.0},"lr_schedule":{"name":"wsd","warmup_steps":312,"decay_fraction":0.1},"training":{"tokens_per_parameters":160.0,"epoch_tokens_per_parameters":0.5,"batch_tokens":131072,"micro_batch_size":16,"checkpoint_policy":"final","checkpoint_epochs":[1],"save_epoch_training_state":true,"log_every":20},"evaluation":{"subset_blocks":1024,"batch_size":128,"seed":12345},"runtime":{"seed":44,"deterministic":false,"device":"cuda:0","amp":true,"compile":true,"compile_mode":"default","sdpa_backend":"auto","fused_optimizer":true,"attention_backend":"sdpa","output_dir":"/nobackup/proj/disk/naiss2024-23-12/personal/zesen/repo/tiny-llm/runs/packed8-awc-wsd-20m-horizon-tuning/experiments/horizon-160/b1-0.95-b2-0.999-lr-0.002-s44/attempt-001","cpu_threads":8},"decentralized":{"num_models":8,"topology":"one_peer_exponential","scheme":"awc","adaptive_consensus":null}},"recipe_identity":"a60fbe6402aa0fa1656a4265d0310d0dcb0fb787524f657bea5e45a6c1ad99c6","loader.blocks":3188096}}}},"audit":{"status":"passed","checkedAt":"2026-09-15T18:55:41.223244+00:00","artifacts":4404,"bytes":293107855511,"curves":594,"continuations":396,"collectorSha256":"76cd372717e39dbd92c30bdf883439c23829e4316518b58aa96b30eedfb5f9f3","limitations":["Stored evaluations were audited, not rerun.","Retired checkpoints are verified through receipts and deletion records; deleted bytes cannot be rehashed."]}}
